THE
BIOLOGICAL BULLETIN
PUBLISHED BY
THE MARINE BIOLOGICAL LABORATORY
Editorial Board
GARY N. CALKINS, Columbia University
E. G. CONKLIN, Princeton University FRANK R. LlLLIE, University of Chicago
E. N. HARVEY, Princeton University CARL R. MOORE, University of Chicago
SELIG HECHT, Columbia University GEORGE T. MOORE, Missouri Botanical Garden
LEIGH HOADLEY, Harvard University T. H. MORGAN, California Institute of Technology
M. H. JACOBS, University of Pennsylvania G. H. PARKER, Harvard University
H. S. JENNINGS, Johns Hopkins University W. M. WHEELER, Harvard University
E. E. JUST, Howard University EDMUND B. WILSON, Columbia University
ALFRED C. REDFIELD, Harvard University
Managing Editor
VOLUME LXI
AUGUST TO DECEMBER, 1931
Printed and Issued by
LANCASTER PRESS, inc.
PRINCE 8C, LEMON STS.
LANCASTER, PA.
11
THE BIOLOGICAL BULLETIN is issued six times a year. Single
numbers, 51.75. Subscription per volume (3 numbers), $4.50.
Subscriptions and other matter should be addressed to the
Biological Bulletin, Prince and Lemon Streets, Lancaster, Pa.
Agent for Great Britain: \Vheldon & Wesley, Limited, 2, 3 and
4 Arthur Street, New Oxford Street, London, \V.C. 2.
Communications relative to manuscripts should be sent to
the Managing Editor, Marine Biological Laboratory, Woods
Hole, Mass., between May 1 and November 1 and to the
Institute of Biology, I >ivinity Avenue, Cambridge, Mass., during
the remainder of the year.
Entered October 10, 1902, at Lancaster, Pa., as second-class matter under
Act of Congress of July 16, 1894.
LANCASTER PRESS, INC.
LANCASTER, PA.
CONTENTS
No. 1. AUGUST, 1931
PAGE
Thirty-third Report of the Marine Biological Laboratory 1
TYLER, ALBERT
The Relation between Cleavage and Total Activation in Arti-
ficially Activated Eggs of Urechis 45
SMITH, GEORGE MILTON
The Occurrence of Melanophores in certain Experimental
Wounds of the Goldfish (Carassius auratus) 73
DICKMAN, ALBERT
Studies on the Intestinal Flora of Termites with reference to
their Ability to Digest Cellulose 85
LUTZ, BRENTON R.
The Innervation of the Stomach and Rectum and the Action
of Adrenaline in Elasmobranch Fishes 93
JOHNSON, GEORGE E., AND NELSON J. WADE
Laboratory Reproduction Studies on the Ground Squirrel,
Citellus tridecemlineatus pallidus, Allen 101
BURKENROAD, M. D.
A New Pentamerous Hydromedusa from the Tortugas 115
ONORATO, A. R., AND H. W. STUNKARD
The Effect of certain Environmental Factors on the Develop-
ment and Hatching of the Eggs of Blood Flukes 120
No. 2. OCTOBER, 1931
PATTERSON, J. T.
Continuous versus Interrupted Irradiation and the Rate of
Mutation in Drosophila 133
TORVIK, M. M.
Genetic Evidence for Diploidism of Biparental Males in Ha-
brobracon 139
WTERNER, ORILLA STOTLEK.
The Chromosomes of the Domestic Turkey 157
ALEXANDER, GORDON
The Significance of Hydrogen Ion Concentration in the Biology
of Euglena gracilis Klebs 165
iii
38883
iv CONTEXTS
REDFIELD, A. C., AND M. FI.ORKIX
The Respiratory Function of the Blood of Urechis caupo . ... 185
SCOTT, \Y. J.
Oxygen and Carbon Dioxide Transport by the Blood of the
Urodele, Amphiuma tridactyla . . . .... 211
MAST, S. O.
Movement and Response in Ditrlugia with special reference
to the Nature of Cytoplasmic Contraction 223
STUNKARD, H. \Y.
The Effect of Dilution of Sea Water on the Activity and Lon-
gevity of Certain Marine Cercariae 242
No. 3. DKCKMUKR, 1931
HARVEY, E. NEWTON
The Tension at the Surface of Marine Eggs, especially those
of the Sea Urchin, Arbacia 273
TAYLOR, G. WKLLFORD, AND E. NEWTON HAKVKY
The Theory of Mitogenetic Radiation . . 280
YYlHTAKKk, D. M.
Some Observations on the Eggs of Fucus and upon their
Mutual Influence in the Determination of the Developmental
Axis 294
COE, WESLEY R.
Spermatogenesis in the California Oyster (Ostrea lurida) .... 309
BLUM. H. F., AND G. C. Me -BRIDE
Studies of Photodynamic Action, III. The difference in mech-
anism between photodynamic hemolysis and hemolysis by
non-irradiated cosine 316
CAROTHERS, E. ELEANOR
The Maturation Divisions and Segregation of Heteromorphic
Homologous Chromosomes in Acrididae (Orthoptera) 324
ADOLPM, EDWARD F.
The Size of the Body and the Size of the Environment in the
Growth of Tadpoles 350
ADOLPH, EDWARD F.
Body Size as a Factor in the Metamorphosis of Tadpoles. . . 376
JAHN, THEO. L.
Studies on the Physiology of the Euglenoid Flagellates, III.
The effect of hydrogen ion concentration on the growth of
Euglena gracilis Klebs 387
HALL, VICTOR E.
The Muscular Activity and Oxygen Consumption of Urechis
• aupo 400
CONTENTS v
BAUMBERGER, J. P., AND L. MIOIAKLIS
The Blood Pigments of Urechis caupo 417
FLORKIN, MARCEL, AND ALFRED C. REDFIELD
On the Respiratory Function of the Blood of the Sea Lion. . . 422
ROOT, R. W.
The Respiratory Function of the Blood of Marine Fishes. ... 427
HALL, F. G.
The Respiration of Puffer Fish 457
TANG, Pi-; i -SUNG
The Rate of Oxygen Consumption of Asterias Eggs before
and after Fertilization • • 468
FAULKNER, G. H.
Notes on the Feeding Mechanism and on Intestinal Respira-
tion in Chaetopterus variopedatus • • 472
WHITING, P. W.
Diploid Male Parts in Gynandromorphs of Habrobracon . . . 478
WHITING, P. W., AND M. F. STANCATI
A Gynandromorph of Habrobracon from a Post-reduced Bi-
nucleate Egg 481
WILLIAMS, MARY MORRISON, AND M. H. JACOBS
On Certain Physiological Differences between Different Prepa-
rations of So-Called "Chemically Pure" Sodium Chloride. . . 485
WELSH, JOHN H.
Specific Influence of the Host on the Light Responses of
Parasitic Water Mites 497
PARPART, ARTHUR K.
Is Osmotic Hemolysis an All-or-None Phenomenon? 500
PARPART, A. K., W. R. AMBERSON AND D. R. STEWART
The Determination of Hemoglobin Concentration in Dilute
Solutions v 518
Vol. LXI, No. 1 August, 1931
THE
BIOLOGICAL BULLETIN
PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY
THE MARINE BIOLOGICAL LABORATORY
THIRTY-THIRD REPORT FOR THE YEAR 1930—
FORTY-THIRD YEAR
I. TRUSTEES AND EXECUTIVE COMMITTEE (AS OK AUGUST 12,
1930) 1
LIBRARY COMMITTEE 3
II. ACT OF INCORPORATION 3
III. BY-LAWS OF THE CORPORATION 3
IV. REPORT OF THE TREASURER 5
V. REPORT OF THE LIBRARIAN 9
VI. REPORT OF THE DIRECTOR 10
Statement 10
Addenda :
1. The Staff, 1930 15
2. Investigators and Students, 1930 17
3. Tabular View of Attendance 28
4. Subscribing and Cooperating Institutions. 1930 .... 28
5. Evening Lectures, 1930 29
6. Shorter Scientific Papers, 1930 31
7. Members of the Corporation 34
I. TRUSTEES
EX OFFICIO
FRANK R. LILLIE, President of the Corporation, The University of Chicago.
MERKEL H. JACOBS, Director, University of Pennsylvania.
LAWRASON RIGGS, JR., Treasurer, 25 Broad Street, New York City.
GARY N. CALKINS, Clerk of the Corporation, and Seer clary of the Board
of Trustees, Columbia University.
EMERITUS
CORNELIA M. CLAPP, Mount Holyoke College.
C. R. CRANE, New York City.
H. H. DONALDSON, Wistar Institute of Anatomy and Biology.
OILMAN A. DREW, Eagle Lake, Florida.
WILLIAM PATTEN, Dartmouth College.
W. B. SCOTT, Princeton University.
E. B. WILSON, Columbia University.
1 1
2 MARIN'E BIOLOGICAL LABORATORY
TO SERVK UXTIL l'>34
E. R. CLARK, University of Pennsylvanin.
E. G. CONKLIN, Princeton University.
OTTO C. GLASER, Amherst College.
Ross G. HARRISON. Yale University.
E. N. HARVEY, Princeton University.
H. S. JENNINGS, Johns Hopkins University.
F. P. KNOWLTOX, Syracuse University.
M. M. METCALF, Johns Hopkins University.
TO SERVE UNTIL 1933
H. C. BRADLEY, University of Wisconsin.
I. F. LEWIS, University of Virginia.
R. S. LILLIE, The University of Chicago.
E. P. LYON, University of Minnesota.
C. E. McCLUNG, University of Pennsylvania.
T. H. MORGAN, California Institute of Technology.
A. C. REDFIELD, Harvard University Medical School.
D. H. TENXENT, Bryn Maur College.
TO SERVF. rxTii. 1032
R. CHAMBERS, Washington Square College. Xc\v York University.
\Y. E. GARREY, Vanclerhilt University Medical School.
CASWELL GRAVE. Washington University.
M. J. GREENMAN. Wistar Institute of Anatomy and Biology.
R. A. HARPER. Columbia University.
A. P. MATHEWS, The University of Cincinnati.
G. H. PARKER, Harvard University.
C. R. STOCKARD, Cornell University Medical College.
TO SERVE UNTIL 1931
H. C. BUMPUS. Broun University.
W. C. CURTIS, University of Missouri.
B. M. DUGGAR, University <>f Wi-con^in.
GEORGE T. MOORE. Missouri Botanical Garden. St. Louis.
W. J. V. OSTERIIOUT, Member of the Rockefeller Institute for Medical
Research.
J. R. SCIIRAMM. University of Pennsylvania.
WILLIAM M. WHEELER. Bussev Institution. Harvard University.
LORANDE L. WOODRUFF, Yale University.
EXECUTIVE COMMITTEE OF THE BOARD OF TRUSTEES
FRANK I\. LILLIE, R.\- Off. Chairman.
MERKEL H. JACOBS, E.r. Off.
LAWRASON RIGGS, JR., l:..v. Off.
G. N. CALKINS, to serve until l'>31.
L. L. WOODRUFF, to serve until 1931.
W. C. CCKTIS, to serve until 1(>32.
A. C. REDFIELD. to serve until 1(>32.
ACT OF INCORPORATION
THE LIBRARY COMMITTEE
C. E. McCLUNG, Chairman.
ROBERT A. BUDINGTON.
E. E. JUST.
M. M. METCALF.
ALFRED C. RF.DFIELD.
A. H. STURTEVANT.
w*wv
^S.-srfe.O.
No. 3170
II. ACT OF INCORPORATION
COMMONWEALTH OF MASSACHUSETTS
Be It Known, That whereas Alpheus Hyatt. William San ford Stevens,
William T. Sedgwick, Edward G. Gardiner, Susan Minns, Charles Sedg-
wick Minot, Samuel Wells, William G. Farlmv, Anna D. Phillips and B.
H. Van Vleck have associated themselves with the intention of forming
a Corporation under the name of the Marine Biological Laboratory, for
the purpose of establishing and maintaining a laboratory or station for scien-
tific study and investigation, and a school for instruction in biology and
natural history, and have complied with the provisions of the statutes of this
Commonwealth in such case made and provided, as appears from the cer-
tificate of the President, Treasurer, and Trustees of said Corporation, duly
approved by the Commissioner of Corporations, and recorded in this office;
Now, therefore, I, HENRY B. PIERCE, Secretary of the Commonwealth
of Massachusetts, do hcreb\ certify that said A. Hyatt, W. S. Stevens,
W. T. Sedgwick, E. G. Gardiner. S. Minns, C. S. Minot. S. Wells, W.
G. Farlow, A. D. Phillips, and B. H. Van Vleck, their associates and suc-
cessors, are legally organized and established as, and are hereby made, an
existing Corporation, under the name of the MARINE BIOLOGICAL
LABORATORY, with the powers, rights, and privileges, and subject to
the limitations, duties, and restrictions, which by law appertain thereto.
Witness my official signature hereunto subscribed, and the seal of the
Commonwealth of Massachusetts hereunto affixed, this twentieth day of
March, in the year of our Lord One Thousand Eight Hundred and Eighty-
Eight.
[SEAL]
HENRY B. PIERCE,
Secretary of the Commonwealth.
III. BY-LAWS OF THE CORPORATION OF THE MARINE
BIOLOGICAL LABORATORY
I. The annual meeting of the members shall be held on the second
Tuesday in August, at the Laboratory, in Woods Hole, Mass., at 12 o'clock
noon, in each year, and at such meeting the members shall choose by ballot
a Treasurer and a Clerk to serve one year, and eight Trustees to serve
four years. There shall be thirty-two Trustees thus chosen divided into
four classes, each to serve four vears, and in addition there shall be two
4 MARINE BIOLOGICAL LABORATORY
groups of Trustees as follows: (a) Trustees ex officio, who shall be the
President of the Corporation, the Director of the Laboratory, the Associate
Director, the Treasurer and the Clerk; ( b} Trustees Emeritus, who shall be
elected from the Trustees by the Corporation. Any regular Trustee who
has attained the age of seventy years shall continue to serve as Trustee
until the next annual meeting of the Corporation, whereupon his office as
regular Trustee shall become vacant and be filled by election by the Cor-
poration and he shall become eligible for election as Trustee Emeritus for
life. The Trustees i\r officio and Emeritus shall each have the same right
to vote as the regular Trustees.
The Trustees and officers shall hold their respective offices until their
successors are chosen and have qualified in their stead.
II. Special meetings of the members may be called by the Trustees to
be held in Boston or in Woods Hole at such time and place as may be
designated.
III. The Clerk shall give notice of meetings of the members by pub-
lication in some daily newspaper published in Boston at least fifteen days
before such meeting, and in case of a special meeting the notice shall state
the purpose for which it is called.
IV. Twenty-five members shall constitute a quorum at any meeting.
V. The Trustees shall have the control and management of the affairs
of the Corporation; they shall present a report of its condition at every
annual meeting; they shall elect one of their number President of the Cor-
poration who shall also be Chairman of the Board of Trustees; they shall
appoint a Director of the Laboratory ; and they may choose such other officers
and agents as they may think best; they may fix the compensation and
define the duties of all the officers and agents; and may remove them, or
any of them, except those chosen by the members, at any time ; they may
fill vacancies occurring in any manner in their own number or in any of
the offices. They shall from time to time elect members to the Corporation
upon such terms and conditions as they may think best.
VI. Meetings of the Trustees shall be called by the President, or by
any two Trustees, and the Secretary shall give notice thereof by written
or printed notice sent to each Trustee by mail, postpaid. Seven Trustees
shall constitute a quorum for the transaction of business. The Board of
Trustees shall have power to choose an Executive Committee from their
own number, and to delegate to such Committee such of their own powers
as they may deem expedient.
VII. The accounts of the Treasurer shall he audited annually by a
certified public accountant.
VIII. The consent of every Trustee shall be necessary to dissolution
of the Marine Biological Laboratory. In case of dissolution, the property
shall be disposed of in such manner and upon such terms as shall be de-
termined by the affirmative vote of two-thirds of the Board of Trustees.
IX. These By-laws may be altered at any meeting of the Trustees, pro-
vided that the notice of such meeting shall state that an alteration of the
By-laws will be acted upon.
X. Any member in good standing may vote at any meeting, either in
person or by proxy duly executed.
REPORT OF THE TREASURER 5
IV. THE REPORT OF THE TREASURER
To THE TRUSTEES OK THE MARINE BIOLOGICAL LABORATORY :
Gentlemen: Herewith is submitted my report as Treasurer of the
Marine Biological Laboratory for the year 1930.
The accounts have been audited by Seamans, Stetson and Tuttle,
certified public accountants. A copy of their report is on file at the
Laboratory and is open to inspection by members of the Corporation.
At the end of the year 1930, the book value of the General Endow-
ment Fund in the hands of the Central Hanover Bank and Trust Com-
pany (of New York) as Trustee was $908,915 in securities and $34.50
in cash. The actual market value of the securities in this fund on the
9th day of May taking the mortgages at face value, was $931,981.25,
showing a very satisfactory appreciation of the value over cost.
The book value of the Library Fund was $199,922.50 in securities
and $77.50 in cash. The actual market value of the securities on May
9th was $203, 03 1.25.
At the end of the year the Lucretia Crocker Fund consisted of securi-
ties of the book value of $4,707.59 and $374.32 in cash.
The Bio Club Scholarship Fund consisted of a mortgage participation
of $2,000 and cash of $31.28, the Reynold A. Spaeth Memorial Lecture
Fund of $3,000 in mortgage securities and cash of $75.46.
The Reserve Fund, consisting of the proceeds of the sale of part of
the Bar Neck property to the Woods Hole Oceanographic Institution,
consisted at the end of the year of bonds of the book value of $20,868.75
and cash of $3,090.55, of which cash $3,000 was later paid out under
the contract, leaving net proceeds of the transaction of $20,959.30 which
is being held with its income to meet maturing mortgage obligations or
for such other purposes as the Trustees may decide.
The Retirement Fund at the end of the year consisted of $15,800 in-
vested in mortgage participations, less an overdraft of $9.73, leaving
$15,790.27.
The land, buildings, equipment and library, excluding the Devil's
Lane and Gansett property, represented an investment of $1,617,086.71,
less depreciation of $246,625.64, or a net amount of $1,370,461.07.
Current expenses including depreciation exceeded income for the
year by $3,767.25.
Over $29,000 was expended from current funds on buildings, equip-
ment and library.
At the end of the year the Laboratory owed $1,640.99 on accounts
payable and $27,000 on bonds secured by mortgage.
Following is the Balance Sheet as of December 31, 1930, and the
condensed statement of income and outgo for the year, also the surplus
account.
6 MARINE BIOLOGICAL LABORATORY
EXHIBIT A
MARINE BIOLOGICAL LABORATORY BALANCE SHEET,
DECEMBER 31, 1930
Assets
Endowment Assets and Equities :
Securities and Cash in Hands of Central Hanover
Bank & Trust Company (of New York)
Trustee— Schedules I-a and I-b $1,108,949.50
Securities and Cash — Minor Funds-
Schedule II . 10.188.65 $1,119.138.15
Plant Assets :
Land— Schedule IV $ 97,103.05
Buildings— Schedule IV 1.207.354.03
Equipment— Schedule IV 155,222.28
Library— Schedule IV 157.407.35 $1. (.17.086.71
Less Reserve for Depreciation 246,625.64
$1.370.461.07
Securities and Cash in Reserve Fund 23.675.43
Cash in Dormitory Buildinu Fund SIS.% $1,394.955.46
Current Assets :
Cash ? 18.010.39
Accounts — Receivable 18.902.69
Inventories :
Supply Department $ 29.063.54
Bulletin 7,951.85 37.015.39
Investments :
Devil's Lane Property $ 37.780.91
Gansett Property 2.273.34
Stock in General Biological
Supply IlmiM'. Inc 12700.00
Retirement Fund Assets . 15.790.27 68.544.52
Prepaid Insurance 3.992.51 $146.465.50
Liabilities
Endowment Funds :
General Endowment Funds— Schedule III $1.108.949.50
Minor Endowment Funds— Schedule III 10.188.65 $1.119.138.15
Plant Funds :
Donations and Gifts— Schedule III $1.025.548.61
Other Investments in Plant from Gifts and Cur-
rent Funds 364.406.85
$1,389.955.46
Mortgage. Danchakoff Estate 2.000.00
Accrued Charges on Sale of Bar Neck Land 3.000.00 $1.394.955.46
REPORT OF THE TREASURER
Current Liabilities and Surplus :
Mortgage, Devil's Lane Property $ 25,000.00
Accounts— Payable 1,640.99
Woods Hole Oceanographic Institution:
Amount received for Purchase of
Books for their Library $2,500.00
Less Expenditures 2,147.07 352.93
Items in Suspense ( Net) 70.49
$ 27,064.41
Current Surplus— Exhibit C 119,401.09 $146,465.50
EXHIBIT B
MARINE BIOLOGICAL LABORATORY INCOME AND EXPENSE.
YEAR ENDED DECEMBER 31, 1930
Total Net
Expense Income Expense Income
Income :
General Endowment Fund .... $ 48,020.46 $ 48,020.46
Library Fund 9,270.24 9,270.24
Gifts 500.00 500.00
Instruction 8.110.03 10,230.00 2,119.97
Research 4.069.37 16,261.06 12,191.69
Evening Lectures 135.48 135.48
Biological Bulletin and Member-
ship Dues 7.557.17 9,421.18 1,864.01
Supply Department-
Schedule V 62.030.00 62,162.82 132.82
Mess— Schedule VI 30.943.36 32,973.28 2,029.92
Dormitories —
Schedule VII 31.188.42 13,764.28 17,424.14
( Interest and Depreciation
charged to above three De-
partments. See Schedules
V. VI. and VII ) 35,424.79 35,424.79
Dividends, General Biological
Supply House. Inc 2,540.00 2,540.00
Rent, Danchakoff Cottages .... 634.11 1,039.00 404.89
Rent, Microscopes 462.50 462.50
Rent, Garage, Railway, etc. ... 154.90 154.00
Rait, Newman Cottage 137.27 150.00 12.73
Rent, Janitor's House 35.84 422.50 386.66
Sale of Duplicate Library Sets 2,198.13 2.198.13
Interest on Bank Balances" 529.87 529.87
Sundry Items 10.64 10.64
Maintenance of Plant :
New Laboratory Expense 16,839.26 16,839.26
Chemical and Special Appa-
ratus 10.783.01 10,783.01
Maintenance, Buildings and
Grounds 9,892.82 9.892.82
MARINE BIOLOGICAL LABORATORY
Library Department Expenses 8,912.66 8,912.66
Carpenter Department Ex-
penses 1,516.91 1,516.91
Truck Expenses 851.48 851.48
Sundry Expenses 772.69 772.69
Bar Neck Property Expenses- 162.54 162.54
Workmen's Compensation In-
surance 592.59 592.59
General Expenses :
Administration Expenses .... 14.509.68 14.509.68
Endowment Fund Trustee . . . 787.50 787.50
Interest on Loans 120.00 120.00
Bad Debts 317.98 317.98
Naples Zoological Station, for
Research 250.00 250.00
Mosquito Fund Contribution 100.00 100.00
Reserve for Depreciation 38,052.73 38.052.73
Excess of Expenses over Income
carried to Current Surplus-
Exhibit C 3.767.25 3767.25
$213,878.11 $213,878.11 $122,021.47 $122.021.47
EXHIBIT C
MARINE BIOLOGICAL LABORATORY, CURRENT SURPLUS ACCOUNT
YEAR KNMKD DECEMBER 31, 1930
Balance, January 1. 1930 $119,933.29
Add:
Reserve for Depreciation charged to Plant Funds 38 052.73
Income from Retirement Fund 603.30
Cash received from Sale of Plant Assets deposited in Current
Cash (Motor-Boat) 50.00
$158.839.32
Deduct:
Payments from Current Funds during Year for Plant Assets as
shown in Schedule IV,
Buildings $ 479.17
Equipment 5.464.23
Library Books, etc 23,099.38
$29;042.78
Purchase of Books from Balance of General Educa-
tion Board Gift of $50,000.00 for Purchase of
Books 5.708.20
Payment of Pensions from Retirement Fund 720.0C
Excess of Expenses over Income for Year as shown
in Exhibit B . 3,767.25 39,238.23
Balance. December 31, 1930— Exhibit A $119.401.09
Respectfully submitted,
LAWRASON RIGGS. JR..
Treasurer.
KKI'ORT OF TIIK LIBRARIAN
V. THE KKI'ORT OF THE LIBRARIAN
The important feature of W30, \vhich was the establishment of a
regular endowment fund for the Library which would ordinarily give,
along with the usual laboratory allowance, about $24,000 annually, was
included in the report of last year. A general statement of the future
apportionment of this sum as there given has been carried out in fact.
A very important addition occurred in the spring, however, when the ,
Director of the Woods Hole Oceanographic Institution placed $5,000
at the disposal of the Library to be used exclusively for the purchase of
oceanographic books and journals. Of this amount, $2,149.73 had been
spent at the end of the year 1930. The items thus purchased are indi-
cated specifically in the general statement of additions to the Library as
follows : journal subscriptions were 346, 24 new, and of these 5 were
for the Woods Hole Oceanographic Institution. One hundred and fif-
teen books were purchased, 45 for oceanography. Back sets of journals
were filled in complete to the number of 45, and 37 only partially com-
pleted— none of these were regarded as strictly for the Woods Hole
Oceanographic Institution. The number of journals received in ex-
change for the Biological Bulletin was 442, an increase of 22, and 15
back sets that we needed were filled in. The reprint collection was
augmented by 5,573.
The Library consists, then, of 26,519 bound journal volumes, 4,991
books, 64,231 reprints; and is receiving 1,060 current journals.
Gifts of books have been made to the Library by the following pub-
lishers, and the Librarian takes this opportunity to acknowledge these in
the name of the Marine Biological Laboratory Library, although formal
thanks have in all cases been directly addressed by letter.
P. Blakiston's Son & Co 9
R. R. Bowker Co 1
Chemical Foundation, Inc 1
Chicago University Press 5
Harvard University Press 2
Paul B. Hoeber 2
Henry Holt & Co 1
Alfred A. Knopf 2
J. B. Lippincott Co 1
McGraw-Hill Book Co., Inc 5
Macmillan Co 21
C. V. Mosby Co 1
W. B. Saunders Co 4
Wm. Wood & Co 2
Yale University Press 1
10 MARINE BIOLOGICAL LABORATORY
VI. THE REPORT OF THE DIRECTOR
To THE TRUSTEES OF THE MARINE BIOLOGICAL LABORATORY :
Gentlemen: I beg to submit herewith a report of the forty-third ses-
sion of the Marine Biological Laboratory for the year 1930.
1. Attendance. The attendance for 1930 showed a slight increase
over that of the preceding year in the numbers of both investigators and
students, the figures for 1930 being 337 investigators and 136 students
as compared with 329 investigators and 125 students in 1(>29. An in-
spection of the Tabular View of .Attendance on page 28 will show ihat
since 1927. when for the first time the research rooms in both the brick
and the wooden buildings were practically all in use at the same time,
the number of independent investigators has scarcely changed, except
for the record-breaking summer of 1929 when visiting foreign physi-
ologists, most of whom came to the Laboratory after the crowded sea-
son, swelled the total to figures not likely to lit1 reached under normal
conditions. On the other hand, investigators under instruction and re-
search assistants whose numbers are not limited by that of the smaller
laboratories have increased from 85 to 120 in the same period. The
limit for the further increase of this class of investigators, however, is
now in sight, and at the present rate will soon be reached.
The number of students, being limited by action of the Trustees, has
shown only minor fluctuations for many years. The slight falling-off in
1929, caused chiefly by the change in that year in the times for holding
the courses, was more than made up in 1930, though the maximum num-
ber which may at present be admitted to the courses, namely 142. has
not yet been reached. This failure of the registration to reach its maxi-
mum value is not due to a deficiency of applications, since in nearly all
of the courses the number of applicants greatly exceeds the number of
available places, but rather to late withdrawals of students who have been
accepted. To discourage such withdrawals, which are unfair to the
rest of the applicants, who have usually in the meantime made other ar-
rangements for the summer, the Executive Committee has recently voted
to make a substantial increase in the registration fee which is forfeited
in case of withdrawal.
Following the custom of the past three years, there are here pre-
sented figures which show the distribution of the attendance of investi-
gators throughout the four seasons, including that of 1930. for which
the necessary records have been kept.
REPORT OF THE DIRECTOR
11
1927
1928
1929
1930
May
30
7
15
9
6
June
10
50
64
55
50
• •
20
' 114
140
139
153
ft
30
?] •>
240
197
208
lulv
10
247
281
238
253
• •
20
247
282
242
250
a
30
245
272
249
253
August
10
234
250
256
254
ti
20
208
226
243
245
.4
30
168
183
220
204
September
10
110
112
157
122
tt
20
50
43
59
44
,i
30 .
12
14
14
8
ujLIBRAKY
2. The Report of the Treasurer. This report shows that the total
assets of the Laboratory at the end of 1930 were $2.660.559.11 as com-
pared with $2,660,478.82 at the end of 1929. A further analysis of the
figures shows that the hook value of the endowment fund has remained
practically stationary (though the Treasurer calls attention to a grati-
fying appreciation in the market value of the securities represented)
while additions to the plant assets, chiefly in the form of hooks and new
equipment, have about balanced the estimated depreciation on buildings
and equipment. A decrease in the value of the land held by the Lab-
oratory and the simultaneous appearance of a reserve fund of approxi-
mately twenty thousand dollars are accounted for by the sale by the
Laboratory to the Woods Hole Oceanographic Institution of the tract
of land upon which the new building of the latter institution now stands.
The income of the Laboratory increased from $200,408.91 in 1929
to $210,110.86 in 1930. A part of this increase is of a temporary na-
ture only, as for example that from the sale of duplicate sets by the
library; of the remainder the largest single item is the additional in-
come from the funds appropriated last year by the General Education
Board. In 1930 for the first time the full annual income from this fund
became available. Since, however, this fund is at present being ap-
plied exclusively to the support of the library, the income available for
general laboratory purposes remains practically unchanged.
The ordinary expenses of the Laboratory during 1930 showed a
comparatively small increase over those for 1929, but expenses incidental
to the reorganization of the Supply Department and the creation of a
Museum, and, in particular, the reduction of the inventory of the Supply
Department by discarding material originally valued at several thousand
dollars, but for which there is at present little or no sale, have again pre-
vented the appearance of a favorable balance, the excess of expenses
over income after making allowance for depreciation being $3,767.25 as
12 MARINE BIOLOGICAL LABORATORY
compared with the similar deficit on paper of $855.33 for 1929. It is
evident, however, when account is taken of the unusual expenses of the
Supply Department in 1930 and of the necessarily large allowances for
depreciation, that the finances of the Laboratory are in a very sound
condition.
In 1930, for the first time in many years, the sum paid to the Lab-
oratory for research space, chiefly by subscribing and cooperating insti-
tutions, showed a slight decrease. This was to have been expected in
view of business conditions, which have materially reduced the incomes
of most of the institutions concerned. It is a very encouraging fact,
however, that the decrease even under these abnormal conditions
amounted to less than three per cent.
3. The Report of the Librarian shows a continuation of the steady
growth of the library which has been made possible especially by the
generous support of the General Education Board. For purposes of
comparison the figures for 1930 may be added to those listed in the
Director's Report for 1929.
1925 1926 1927 192S 1929 1930
Serials received currently 500 628 764 S74 985 1060
Total number of bound
volumes 15000 18200 22800 2o50i> 28300 31500
Reprints 25000 38000 43000 51000 59000 64000
One especially noteworthy feature of the past year lias been the strength-
ening of the part of the library devoted to the subject of oceanography,
the development in this direction having been made possible by a co-
operative arrangement with the Woods Hole Oceanographic Institution,
assisted by special funds appropriated by that institution.
4. Lectures and Scientific Meetings. During the season of 1930
the number of formal scientific lectures, including the Reynold A.
Spaeth Memorial Lecture delivered by Professor Hardolph Wastenys
of the University of Toronto, was thirteen, with several other evenings
devoted to non-scientific lectures and motion pictures. In addition,
there were held 13 less formal meetings at which 56 shorter papers,
whose titles are given on pages 31 to 33, were presented and discussed.
Two of these meetings were of especial interest. The first, held on
June 27, assumed the character of a celebration of the sixtieth birthday
of Dr. Frank R. Lillie and of the fortieth consecutive year of his con-
nection with the Marine Biological Laboratory. In addition to the
scientific papers presented on that occason. which were all based upon
work having its inception in Dr. Lillie's laboratory, a special address of
congratulation was delivered by 1 )r. I*".. 15. Wilson, and a ship's clock, the
gift of Dr. Lillie's former students, was presented to him by Dr. L. Y.
REPORT OF THK DIRECTOR
Heilbrunn. The evening was concluded by an informal reception at the
M. B. L. Club. The second meeting of an unusual character was the
one held on the morning of July 26 at which 12 papers in the field of
neuro-muscular physiology were presented and discussed. This meet-
ing formed the most important part of a two days' program, social as
well as scientific in character, which was arranged by the workers in this
field and was attended not only by laboratory workers but by a number
of physiologists from a distance. So successful was this meeting that it
is to be hoped that similar ones, devoted to various fields of biological
research may be held in the future.
5. Supply Department and Museum. At the last annual meeting of
the Board of Trustees it was voted to develop for the use of investi-
gators and students working in Woods Hole a museum in which speci-
mens of the local fauna and flora may at all times be available for pur-
poses of study, and in which there may at the same time be preserved
full records of the distribution of all the local forms, the seasons of their
maximum abundance, their breeding habits, etc. The development of
this important activity of the Laboratory was very appropriately placed
in the hands of Mr. George M. Gray, whose long experience as Curator
of the Supply Department has given him unique qualification for such
a position. In order to fill the vacancy thus created in the Supply De-
partment and to provide for a possible ultimate separation of the two
present functions of this Department, namely, that of supplying living
material for experimental purposes to workers at the Laboratory and of
furnishing preserved material to schools and colleges, the General Bio-
logical Supply House of Chicago was invited to assume its temporary
management. In preparation for the new arrangement, Dr. D. L.
Gamble, representing this firm, spent several months in residence in
Woods Hole during the summer of 1930 and has since continued the
general supervision of this Department from Chicago with very satis-
factory results, being ably assisted by Mr. James Mclnnis as Resident
Manager.
6. Facilities for Work witJi X-rays. During the summer of 1930,
through an appropriation of $500.00 by the Committee on the Effects
of Radiation upon Living Organisms of the National Research Council
and with the active assistance of the Chairman of that Committee, Pro-
fessor W. C. Curtis, special facilities, not hitherto available for work
with X-rays and other radiations were provided for workers at the Lab-
oratory. In particular, there was made available throughout the summer
the expert advice of Dr. G. Failla of the Memorial Hospital, New York,
together with the assistance of competent technicians in the operation of
the apparatus. Several manufacturers also furnished very valuable aid
14 MARIXE BIOLOGICAL LABORATORY
of various sorts which is here gratefully acknowledged. So successful
was this arrangement in 1930 that it is gratifying to he ahle to announce
at the time of the writing of this report that it will he continued and
somewhat extended in 1931.
7. The Woods Hole Oceanographic Institution. The position which
Woods Hole has held for many years as one of the leading hiological
centers in the United States was materially strengthened hy the erection
during the past year of the large and splendidly equipped building which
will be the permanent headquarters of the \Yoods Hole Oceanographic
Institution. The land on which this building stands was formerly held
by the Marine Biological Laboratory and leased by it to the liar Xeck
Corporation. At a special meeting of the Board of Trustees, held in
Washington. I). C., on April 28, 1930. it was voted to enter into an
agreement with the Woods Hole Oceanographic Institution and the
liar Neck Corporation by which the former institution would acquire
by purchase1 from the Marine Biological Laboratory approximately
54,000 square feet of the westerly portion of the so-called " Bar Xeck
Wharf." Full details as to this agreement will be found in the Re-
port of the Auditors for 1930. Though there is no official connection
between the Woods Hole Oceanographic Institution and the Marine
Biological Laboratory, the work of each institution will supplement that
of the other, and it is planned that there shall be close scientific co-
operation between them. This cooperation has already assumed the
form of a sharing of library facilities and Mess accommodations and
will be extended in the future in all possible ways.
8. The IJoanl of Trustees. ( )ne change in the Board of Trustees
occurred during the past year, Professor William Patten of Dartmouth
College having been elected Trustee Emeritus at the annual meeting of
the Corporation and Professor E. R. Clark of the Cniversity of Penn-
sylvania having been selected to till tin- place thereby made vacant in
the Class of 1934.
9. Gifts. Appreciative acknowledgment is made of the gift by Mr.
Ware Cattell and the " Collecting Net " of $500.00 for scholarships to
students who in the courses given bv the Laboratory show unusual
promise as research workers.
There are appended as parts of this report :
1. The Staff, 1930.
2. Investigators and Students, 1930.
3. A Tabular View «>f Attendance. 1926-1930.
4. Subscribing and Cooperating Institutions, 1930.
5. Evening Lectures, 1930.
REPORT OF THE DIRECTOR
6. Shorter Scientific Papers. 1930.
7. Members of the Corporation, August, 1930.
Respectfully submitted,
M. H. JACOBS,
Director.
1. THE STAFF, 1930
MERKEL H. JACOBS. Director, Professor of General Physiology, University
of Pennsylvania.
Associate Director: -
ZOOLOGY
I. IXVKSTIGATIOX
GARY N. CALKINS, Professor of Protozoology, Columbia University.
E. G. CONKLIN, Professor of Zoology, Princeton University.
CASWELL GRAVE, Professor of Zoology, Washington University.
H. S. JENNINGS, Professor of Zoology, Johns Hopkins University.
FRANK R. LILLIE, Professor of Embryology, University of Chicago.
C. E. McCLUNG, Professor of Zoology, University of Pennsylvania.
S. O. MAST, Professor of Zoology, Johns Hopkins University.
T. H. MORGAN, Director of the Biological Laboratory, California Institute
of Technology.
G. H. PARKER, Professor of Zoology, Harvard University.
E. B. WILSON, Professor of Zoology, Columbia University.
LORANDE L. WOODRUFF. Professor of Protozoology, Yale University.
II. INSTRUCTION
J. A. DAWSON, Assistant Professor of Biology, College of the City of New
York.
T. H. BISSONNETTE, Professor of Biology, Trinity College.
E. C. COLE, Associate Professor of Biology, Williams College.
0. E. NELSEN, Instructor in Zoology, University of Pennsylvania.
A. W. POLLISTER, Instructor in Zoology, Columbia University.
L. P. SAYLES, Instructor in Biology, College of the City of New York.
A. E. SEVERINGHAUS, Assistant Professor of Anatomy, College of Phy-
sicians and Surgeons, Columbia University.
JUNIOR INSTRUCTORS
B. R. COONFIELD, Professor of Biology, Southwestern College.
1. B. HANSEN, Graduate Student, University of Chicago.
PROTOZOOLOGY
I. INVESTIGATION
(See Zoology}
II. INSTRUCTION
GARY N. CALKINS, Professor of Protozoology, Columbia University.
RACHEL BOWLING, Instructor in Zoology, Columbia University.
ROBERTS RUGH, Assistant in Zoology. Columbia University.
16 MARINE BIOLOGICAL LABORATORY
EMBRYOLOGY
I. INVESTIGATION
(Sec Zoology}
II. INSTRUCTION
HUBERT B. GOODRICH, Professor of Biology, Wesleyan University.
BENJAMIN H. GRAVE, Professor of Biology, De Pamv University.
CHARLES PACKARD, Assistant Professor of Zoology, Institute of Cancer
Research, Columbia University.
CHARLES G. ROGERS, Professor of Comparative Physiology, Oberlin College.
B. C. T WITTY, Instructor in Zoology, Yale University.
PHYSIOLOGY
I. INVESTIGATION
HAROLD C. BRADLEY, Professor of Physiological Chemistry, University of
Wisconsin.
WALTER E. CARREY, Professor of Physiology, Vanclerbilt University Med-
ical School.
RALPH S. LILLIE, Professor of General Physiology, University of Chicago.
ALBERT P. MATHEWS, Professor of Biochemistry, University of Cincinnati.
II. INSTRUCTION
Teaching Staff
WILLIAM R. AMBERSON, Assistant Professor of Physiology, University of
Pennsylvania.
PHILIP BARD, Assistant Professor of Physiology, Princeton University.
HALLOWELL DAVIS, Assistant Professor of Physiology, Harvard University.
RALPH W. GERARD, Assistant Professor of Physiology, University of Chi-
cago.
CHARLOTTE HAYWOOD, Assistant Professor of Physiology, Vassar College.
LEONOR MICHAELIS, Member of the Rockefeller Institute, New York City.
Special Lecturers
EDWIN J. COHN, Associate Professor of Physical Chemistry, Harvard Uni-
versity.
HENRY J. FRY, Associate Professor of Biology, Washington Square College,
New York University.
E. NEWTON HARVEY, Professor of Physiology, Princeton University.
SELIG HECHT, Professor of Biophysics, Columbia University.
MI-.RKEL H. JACOBS, Professor of General Physiology, University of Penn-
sylvania.
BALDUIN LUCKE, Associate Professor of Pathology, University of Pennsyl-
vania.
BOTANY
I. INVESTIGATION
B. M. DUGGAR, Professor of Physiological and Economic Botany, University
of Wisconsin.
C. E. ALLEN, Professor of Botany, University of Wisconsin.
REPORT OF THE DIRECTOR 1 '
S. C. BROOKS, Professor of Zoology, University of California.
IVEY F. LEWIS, Professor of Biology, University of Virginia.
WM. J. ROBBINS, Professor of Botany, University of Missouri.
II. I: xs'i RUCTION
WILLIAM RANDOLPH TAYLOR, Professor of Botany, University of Penn-
sylvania.
H. C. BOLD, Instructor in Botany, University of Vermont.
JAMES P. POOLE, Professor of Evolution, Dartmouth College.
LIBRARY
PRISCILLA B. MON -IGOMKRY ( MRS. THOMAS H. MONTGOMERY, JR.), Li-
brarian.
DEBORAH LAWRENCE. Secretary.
HESTER ANN BRADBURY, LILLIAN F. BRIGGS, MARY A. ROHAN, Assistants.
CHEMICAL SUPPLIES
OLIVER S. STRONG. Professor of Xeurology and Xeuro-Histology. Columbia
University, Chemist.
APPARATUS ROOM
SAMUEL E,. POND, Assistant Professor of Physiology, Medical School, Uni-
versity of Pennsylvania, Custodian of Apparatus.
MUSEUM
GEORGE M. GRAY, Curator.
SUPPLY DEPARTMENT
JAMES MC!NNIS, Manager. WALTER KAHLER, Collector.
A. M. HILTON, Collector. GEOFFREY LEHY, Collector.
MILTON B. GRAY, Collector. A. W. LEATHERS, Shipping.
BOATS
JOHN J. VEEDER, Captain. E. M. LEWIS, Chief Engineer.
F. M. MACNAUGHT, Business Manager.
HERBERT A. HILTON, Superintendent of Buildings and Grounds.
THOMAS LARKIN. Superintendent of Mechanical Department.
LESTER F. Boss, Mechanician.
J. D. GRAHAM, Glass-blowing Service.
A. R. APGAR, Photographic Service.
WILLIAM HEMENWAY, Carpenter.
2. INVESTIGATORS AND STUDENTS, 1930
Independent Investigators
ABRAMSON, HAROLD A.. Instructor. Harvard University.
AMBERSON. WILLIAM R.. Assistant Professor of Physiology, University of Penn-
sylvania.
2
18 MARINE BIOLOGICAL LABORATORY
ARMSTRONG, PHILIP B., Instructor in Anatomy, Cornell University Medical College.
ASHWORTH, JAMES H., Professor of Natural History, University of Edinburgh.
ASTROM, I. ELISABETH, Class Assistant, University of Toronto.
BAITSELL, GEORGE A., Professor of Biology, Yale University.
BAKER, HORACE B., Associate Professor, University of Pennsylvania.
BALL, ERIC G., National Research Fellow in Medicine, Johns Hopkins University
Medical School.
BARD, PHILIP, Assistant Professor of Physiology, Princeton University.
BARRON, E. S. GUZMAN, Instructor in Medicine, Johns Hopkins University Medical
School.
BARTH, L. G., National Research Fellow, University of Chicago.
BEAMS, H. W., Dupont Fellow, University of Virginia.
BELKIN, MORRIS, Instructor, New York University.
BIDDLE, RUSSELL L., Teaching Fellow. California Institute of Technology.
BISSONNETTE, T. HUME, Professor of Biology, Trinity College.
BLANCHARD, KENNETH C., Associate Professor, New York University.
BLUMENTHAL, REUBEN, Harrison I-Yllow in Zoology, University of Pennsylvania.
BOLD, HAROLD C.. Instructor in Botany. University of Vermont.
BORODIN, D. N., 621 West 142d Street, New York City, New York.
BOWLING, RACHEL, Instructor in Zoology, Columbia University.
BRAPWAV, WINNIFRED, New York University.
BREITENBECHER, J. K., McGill University.
BRIDGES, CALVIN B., Research Assistant, Carnegie Institution of Washington.
BRONFENBRENNER, J., Professor of Bacteriology, Washington University Medical
School.
BRONK, DETLEV W.. Professor of Biophysics and Director of Johnson Foundation
for Medical" Physics, University of Pennsylvania.
BROOKS, MATILDA M., Research Associate in Biology, University of California.
BROOKS, S. C., Professor of Zoology, University of California.
BURDICK, DONALD L., Instructor in Biology, Union C»lK-m -.
BYTINSKI-SALZ, HANS, Research Fellow, Yale University.
CALKINS, GARY N., Professor of Proto/m'ilogy, Columbia University.
CAROTHERS, E. ELEANOR, Lecturer in Zoology, University of Pennsylvania.
CATTELL, WARE, Research Fellow in Biophysics, Memorial Hospital.
CHALMERS, ELIZABETH, Graduate Assistant. University of Pittsburgh.
CHATTON, EDOUARD, University of Strasbourg, Strasbourg, France.
CHEEVER, CLARENCE A., Boston Society of Natural History, Boston, Mass.
CHIDESTER, FLOYD E., Professor of Zoology, West Virginia University.
CHOUKE, K. S., Assistant Professor of Anatomy. School of Medicine, University
of Colorado.
CHRISTIE, JESSE R., Associate Nematologist, United States Department of Agri-
culture.
CLOWES, G. H. A., Director, The Lilly Research Laboratory. Eli Lilly & Co.
COBB. N. A., Agricultural Technologist and Nematologist, United States Depart-
ment of Agriculture.
COE, W. R., Professor of Biology, Yale University.
COKFEY, J. M., Assistant Bacteriologist, New York State Department of Health.
COLE, ELBF.RT C., Associate Professor of Biology, Williams College.
COLE, KENNETH S., Assistant Professor of Physiology, Columbia University.
COONFIELD, BENJAMIN R., Professor of Biology, Southwestern College.
COOPER, GEORGE O., Instructor, University of Wisconsin.
Coi'ELAND, MANTON, Professor of Biology, Bowdoin College.
COWLES, R. P., Associate Professor of Zodlogy, Johns Hopkins University.
CURTIS, W. C., Professor of Zoology, University of Missouri.
DAVIS, HALLOWELL, Assistant Professor of Physiology, Harvard University Med-
ical School.
REPORT OF THE DIRECTOR 19
DAWSON, ALDEN B., Associate Professor of Zoology, Harvard University.
DAWSON, J. A., Assistant Professor of Biology, College of the City of New York.
DILL, D. B., Assistant Professor of Biochemistry, Harvard University.
DOLLEY. WILLIAM I.., JR., Professor of Biology, University of Buffalo.
Du Bois, DELAFIELD, Washington Square College, New York University.
Du Bois, EUGENE F., Associate Professor of Medicine, Cornell University Med-
ical College.
DuBuissoN, MARCEL, Professor of Zoology, " Ecole des Hautes fitudes," Ghent,
Belgium.
DUGGAR, B. M., Professor of Plant Physiology and Applied Botany, University of
Wisconsin.
EDWARDS, DAYTON J., Associate Professor of Physiology, Cornell University
Medical College.
EDWARDS. H. T., Assistant in Fatigue Laboratory, Harvard University.
FAILLA, G., Physicis-t, Memorial Hospital, New York.
FANKHAUSER, GERHARD, Fellow of the Rockefeller Foundation, University of Chi-
cago.
FINLEY, HAROLD E., Instructor in Zoology, West Virginia State College.
FLORKIN, MARCEL, Research Fellow, Harvard University.
FREW, PRISCILLA E.. Instructor, Hunter College.
FRY, HENRY J., Associate Professor of Biology, Washington Square College,
New York University.
FURTH, JACOB, Associate in Pathology, The Henry Phipps Institute, University of
Pennsylvania.
GARDINER, MARY S., Associate in Biology, Bryn Mawr College.
GARREY, W. E., Professor of Physiology, Vanderbilt University Medical School.
GATES, FREDERICK L., Research Fellow, Harvard University.
GELFAN, SAMUEL, Research Fellow, University of Chicago.
GERARD, R. W., Assistant Professor of Physiology, University of Chicago.
GIBBONS, NORMAN E., Graduate Student, Yale University.
GOLDFORB, A. J., Professor of Biology, College of the City of New York.
GOODRICH, HUBERT B., Professor of Biology, Wesleyan University.
GRAVE, B. H., Professor of Zoology, DePauw University.
GRAVE, CASWELL, Professor of Zoology, Washington University.
GRUNDFEST, HARRY, National Research Council Fellow, Columbia University.
HANCE, ROBERT T., Head of Department of Zoology, University of Pittsburgh.
HARVEY, ETHEL B., Instructor, Washington Square College, New York University.
HARVEY, E. NEWTON, Professor of Physiology, Princeton University.
HAYWOOD, CHARLOTTE, Assistant Professor of Physiology, Vassar College.
HEILBRUNN, L. V., Associate Professor of Zoology, University of Pennsylvania.
HENSHAW, PAUL S., Biophysicist, Memorial Hospital.
HETLER. DONALD M., Instructor in Bacteriology, Washington University Medical
School.
HIBBARD, HOPE, Assistant Professor, Oberlin College.
HILL, SAMUEL E., Assistant in Physiology, Rockefeller Institute.
HINRICHS. MARIE A., Research Associate in Physiology, University of Chicago.
HINTZE, A. LAURA, Assistant Professor of Physiology, Goucher College.
HIRSCH, G. C.. Professor of Zoology, University of Utrecht.
HOPPE, ELLA N., Research Assistant, New York State Department of Health.
HOWE, H. E., Editor, Industrial and Engineering Chemistry.
ROWLAND, RUTH B., Associate Professor of Biology, Washington Square College,
New York University.
HUETTNER, ALFRED F., Associate Professor, Washington Square College, New
York University.
HUGHES, THOMAS P., Associate in Bacteriology, Rockefeller Institute.
HUNTER, LILLIAN M., Graduate Student and Assistant Technician, University of
Toronto.
20 MARINE BIOLOGICAL LABORATORY
HUTCHINSON, G. E., Instructor in Biology. Yale University.
HYMAN. LIBBIE H.. Research Associate, University of Chicago.
JACOBS, M. H., Professor of General Physiology-, University of Pennsylvania.
JENNINGS, H. S.. Professor of Zoology. Johns Hopkins University.
JOHLIN, J. M., Associate Professor of Biochemistry, Vanderbiit University Med-
ical School.
JOHNSON, H. HERBERT, Instructor, College of the City of Xew York.
JUST, E. E., Professor of Zoology, Howard University.
KEIL, ELSA M.. Instructor in Zoology, Rutgers University.
KELTCH, ANNA K.. Research Chemist, Lilly Research Laboratory.
KETTLEKAMP, B. H., Instructor, University of Pittsburgh.
KEVES, D. B., Professor of Industrial Chemistry. University of Illinois.
KEYS. ANCEL B.. Fellow in the Biological Sciences, .National Research Council.
KNOWER. HENRY Me K.. Wistar Institute.
KNOWLTON. FRANK P.. Professor of Physiology, Syracuse University, College
of Medicine.
KOEHRING, VKRA. Beaver College, Jenkintown, Pennsylvania.
KUNITZ, MOSES. Associate Professor. Rockefeller Institute for Medical Research.
LACKEY. J. B.. Professor of Biology. Southwestern University.
LANCEFIELD, D. K.. Associate Professor in Zoology, Columbia University.
LANCEFIELD, REBECCA C. Assistant in Bacteriology, Rockefeller Institute for Medi-
cal Research.
LEWIS, IVEY F., Professor of Biology. University of Virginia.
LII.I.IE. FRANK R.. Chairman of the Department of Zoology, University of Chicago.
I. ii. i. IK, RALPH S., Professor of General Physiology, University of Chicago.
LUCKE, BALDUIN, Associate Professor of Pathology, University of Pennsylvania.
LYNCH. RUTH S.. Instructor in Graduate Zoology. The Johns Hopkins University.
LYON, E. P., Professor of Physics, University of Minnesota.
McCLUNG, C. E., Director, Zoological Laboratory, and Professor of Zoology, Uni-
versity of Pennsylvania.
MAcDouGAi.i., MARY S.. Head of Biology Department. Agnes Scott College.
M< F.WF.N. ROBERT S.. Associate- Professor of Zoology. Oherlin College.
McGLONE. MARTI. is. Instructor in Physiology. University of Pennsylvania.
MANWEIX. RKGIXAI.D I).. Instructor, Johns Hopkins University.
MARSLANP, DOUGLAS A., Assistant Professor of Biology. Washington Square
College, Xew York University.
MATHEWS, A. P.. Professor of Biochemistry. University of Cincinnati.
MATTHEWS. SAMUEI A., Instructor. University of Pennsylvania.
MAVOR, JAMF.S W., Professor of Biology and Head of Department. Union College.
MENKIN, VALY, Fellow in Medicine, Henry Phipps Institute. University of
Pennsylvania.
MEYER. Roi \NI> K.. Research Assistant. University of Wisconsin.
MICHAELIS, LF.ONOR, Member, Rockefeller Institute for Medical Research.
MII.LEH. Hiii'.v M.. Fellow, National Research Council, Johns Hopkins University.
MITCHELL. Piiiiir H., Professor of Physiology. Brown University.
MONNIER, ANTIREE. University of Paris.
MONNIER, A. M.. Assistant at the Sorbonne. Paris.
MORGAN, LILIAN V.. California Institute of Technology.
MORGAN. T. H.. Director of the Biological Laboratory, California Institute of
Technology.
MORGULIS, S.. Professor of Biochemistry. University of Xebraska. College of
Medicine.
MORRILL, C. V., Associate Professor of Anatomy. Cornell University Medical
College.
MORRIS. HEI.F.N S.. Graduate Student. Columbia University.
MULDER, ARTHUR G., Associate Professor of Physiology. University of Tenness-ee.
REPORT OF THE DIRECTOR
NABRIT, S. MILTON, Head of Department of Biology, Morehousc College.
NELSON, OLIN E., Instructor in Zoology, University of Pennsylvania.
NICHOLAS, WARREN W., X-Ray Physicist, National Bureau of Standards.
NONIDEZ, JOSE F., Assistant Professor of Anatomy, Cornell University Medical
College.
PACKARD, CHARLES, Assistant Professor of Zoology, Columbia University.
PARK, ORLANDO, Associate Professor of Biology, Kent State College.
PARMENTER, CHARLES L., Associate Professor of Zoology, University of Penn-
sylvania.
PAYNE, FERNANDAS, Professor of Zoology and Dean of Graduate School, Indiana
University.
PERROT, J. L., Columbia University.
PETRIK, JOSEPH M., Director of the Department of Physiology, Masaryk Uni-
versity.
PHILLIPS, PAUL L., Instructor in Anatomy, Cornell University Medical College.
PHILPOTT, CHARLES H., Lecturer in Medical Zoology, Washington University
Medical School.
PLOUGH, HAROLD H., Professor of Biology, Amherst College.
POLLISTEK, ARTHUR W., Instructor in Zoology, Columbia University.
POND, SAMUEL E., Assistant Professor of Physiology, University of Penn-
sylvania Medical School.
POOLE, JAMES P., Professor of Evolution, Dartmouth College.
POTTER, TRUMAN S., Seymour Coman Fellow, University of Chicago.
RAFFEL, DANIEL, National Research Fellow, Johns Hopkins University.
REDFIELD, HELEN, California Institute of Technology.
REESE, ALBERT M., Head of Department of Zoology, West Virginia University.
RICHARDS. OSCAR W.. Assistant Professor of Biology, Clark University.
ROBERTS, EDITH A.. Chairman of Department of Botany, Vassar College.
ROGERS, CHARLES G., Professor of Comparative Physiology, Oberlin College.
ROMANOFF, ALEXIS L., Research Instructor, Cornell University.
ROOT, \VALTER S., Assistant Professor, Syracuse University Medical School.
RUGH, ROBERTS, Assistant in Zoology, Columbia University.
SAYLES, LEONARD P., Instructor in Zoology, Tufts College.
SCHAUFFLER, WILLIAM G., Princeton, New Jersey.
SCHMIDT, LEON H.. University of Cincinnati.
SCHMITT, FRANCIS O., Assistant Professor of Zoology, Washington University.
SCHRADER, FRANZ, Associate Professor, Bryn Mawr College.
SCHRADER, SALLY-HUGHES, Instructor in Biology, Bryn Mawr College.
SCHULTZ, JACK, Research Assistant, Carnegie Institution of Washington.
SEVERINGHAUS, AURA E.. Associate in Anatomy, Medical School, Columbia Uni-
versity.
SHARMA, B. M., Professor of Anatomy, Tibbi Medical College.
SHOUP, CHARLES S., Assistant Professor of Biology, Vanderbilt University.
SHULL, A. FRANKLIN, Professor of Zoology, University of Michigan.
SHUMWAY, WALDO, Professor of Zoology, University of Illinois.
SICHEL, FERDINAND J., Assistant Instructor, Washington Square College, New
York University.
SLIFER, ELEANOR H., Graduate Student. University of Pennsylvania.
SMITH, FREDERICK, Research Assistant, Rockefeller Institute.
STEINBACH, H. B., Graduate Student, Brown University.
STOCKARD, CHARLES R., Professor of Anatomy, Cornell University Medical College.
STRONG, OLIVER S., Professor of Neurology and Neuro-Histology, Columbia Uni-
versity.
STUNKARD, H. W., Professor of Biology, New York University.
STURDIVANT, HARWELL P., Instructor, Columbia University.
STURTEVANT, A. H., Professor of Genetics, California Institute of Technology.
22 MARINE BIOLOGICAL LABORATORY
SUM WALT, MARGARET, Assistant Professor, Woman's Medical College.
TASHIRO. SHIRO, Professor of Biochemistry, The University of Cincinnati.
TAYLOR, WM. RANDOLPH, Professor of Botany, University of Pennsylvania.
THOMPSON, J. W., Instructor, Swarthmore College.
TOWER, SARAH S.. Instructor in Anatomy, Johns Hopkins University.
TWITTY, VICTOR C., Instructor in Zoology, Yale University.
UHLENHUTH, EDUARD, Professor of Gross Anatomy. University of Maryland
Medical School.
VAN CLEAVE, C. D., Instructor in Anatomy. University of Pennsylvania.
VAN SLYKE, E., Instructor, University of Pittsburgh.
WANG, CHI CHE, 1314 East 56th Street, Chicago, Illinois.
WARREN, HOWARD C., Stuart Professor of Psychology, Princeton University.
WENRICH, D. H., Professor of Zoology, University of Pennsylvania.
WESSON, LAURENCE G., Assistant Professor of Pharmacology, Vanderbilt Univer-
sity Medical School.
WIIKDON, ARTHUR D.. Professor of Zoology and Head of Department, North
Dakota Agricultural College.
WHITAKER, D. M., Assistant Professor of Zoology, Columbia University.
WHITING, ANNA R., Professor, Head of Department of Biology, Pennsylvania
College for Women.
WHITING, P. W., Associate Professor of Zoology, University of Pittsburgh.
WIEMAN, H. L., Professor of Zoology, University of Cincinnati.
WILLIER, B. H., Associate Professor of Zoology. University of Chicago.
WILSON, EDMUND B., DaCosta Professor of Zoology, Emeritus, Columbia Uni-
versity.
WILSON, F. EDWARD, Graduate Student, Clark University.
WILSON, MAY G., Associate, Department of Pediatrics, Cornell University Medical
College.
WOLF, E. ALFRED, Assistant Professor of Zoology. Unm-i-Miy of Pittsburgh.
WOODRUFF, L. L., Professor of Protozoology. Yale University.
WOODS, FARRIS H.. Assistant Profes>or of /mildly. University of Missouri.
YOUNG, WILLIAM C.. Instructor in Biology, Brown University.
ZELENY, CHARLES. Professor of Zoology, University of Illinois.
Beginning Investigators
BAILEY, SARAH Wr., Graduate Student. Radcliffe College.
BROWN, FRANK A., JR., Austin Teaching Fellow, Harvard University.
CALDWELL, LUCILE. Johns Hopkins University.
CHANG, J. H., Graduate Student. University of Chicago.
CHEN, H. T., Student, Harvard University Medical School.
CLINE, ELSIE, Graduate Student, The Johns Hopkins University.
COLDWATER, KENNETH B., Instructor in Zoology, University of Missouri.
COSTELLO, DONALD P., Assistant in Zoology, College of the City of Detroit.
CROASDALE, HANNAH T., Graduate Student, University of Pennsylvania.
CURTIS, MARY ELIZABETH, Graduate Student, Yale University.
DAMBACH, GEORGE J., Graduate Assistant, University of Pittsburgh.
DEARING, WILLIAM H., Graduate Student, University of Pennsylvania.
DEITRICH, JOHN E., Medical Student, Johns Hopkins University.
DIXON, EVELYN C., Graduate Student. Washington University.
DRAPER. JOHN W., Cornell University Medical College.
DREW, WILLIAM, Massachusetts Agricultural College.
DuSnANE, GRAHAM, Wabash College.
ETKIN, WILLIAM, Tutor. College of the City of New York.
FLAMMON. SISTER M. MURIEL. Instructor in Biology, Seton Hill College.
FRENCH, C. S., Harvard University.
GEBRAK, ANTON, Moscow Agricultural Academy.
KKPORT OF THE DIRECTOR
GEIB, DOROTHY A., Student, Johns Hopkins University Medical School.
GEIMAN, QUENTIN M., Graduate Student, University of Pennsylvania.
GENTHER, IDA T., Assistant in Zoology, Washington University.
GRAUBARD, MARC A., Assistant in Zoology, Columbia University.
GREEN, DAVID E., Assistant, New York University.
GUERLAC, HENRY E., Cornell University.
HANSEN, IRA B., Instructor, Wesleyan University.
HAYNES, FLORENCE W., 20 Gorham Road, West Medford, Massachusetts.
HEGNAUER. ALBERT, Assistant in Physiology, University of Rochester, School of
Medicine.
HILEMAN, CLARA M., Teacher of Biology, Columbia University.
HOERR, STANLEY O., Antioch College.
HOOK, SABRA J., Assistant in Zoology, Barnard College, Columbia University.
IGLAUER, CHARLES A., Graduate Student, University of Pennsylvania.
IMLAH, HELEN- W., Graduate Student, Radcliffe College.
JEFFERY, HELEN, Research, Washington University Medical School.
KALTREIDER, NOLAN L., Student, Johnson Foundation, University of Pennsylvania.
KATZ, JACOB D., Assistant Instructor, Washington Square College, New York Uni-
versity.
KERR, THOMAS, Instructor in Biology, New York University.
KIXNEY, ELIZABETH T., Assistant at Barnard College, Columbia University.
KINSBERGEN, MAURICE, Assistant, New York University.
KLOSE, THEODORA, Instructor in Botany, Vassar College.
LEE, KATY, Graduate Assistant in Zoology, University of Missouri.
LHERISSON, CAMILLE, Professor of Biology, University of Haiti Medical School.
LORBERBLATT, ISAAC, Student, Washington University Medical School.
McGouN, RALPH C, JR., Instructor in Biology, Amherst College.
MACMURRAY, MARY T., 8629 109th Street, Richmond Hill. New York.
MACKMULL, GULDEN, Demonstrator of Histology and Embryology, Baugh Insti-
tute of Anatomy.
MILLER, EVELYN H., Graduate Student, University of Pennsylvania.
MILLER, FORREST W., Graduate Assistant, University of Pittsburgh.
MONAGHAN, BETTY R., Assistant, Washington University.
MORRIS, SAMUEL, Instructor in Zoology, University of Pennsylvania.
NELSON, PHYLLIS M., Washington University.
PARPART, ARTHUR K., Instructor and Graduate Student, University of -Pennsylvania.
PARPART, ETHEL R., Technician, University of Pennsylvania.
PITTS, ROBERT F., Student Assistant, Johns Hopkins University.
RANKIN, DOUGLAS, Johns Hopkins University.
ROBERT, NAN L., Instructor, Hunter College.
SANTOS, FELIX V., Graduate Student, The University of Chicago.
SAVIN, MARION B., Graduate Student, University of Pennsylvania.
SCARBOROUGH, J. ELLIOTT, JR., Student of Medicine, Harvard University.
SCHECHTER, VICTOR, Tutor in the Department of Biology, College of the City of
New York.
SCHWEITZER, MORTON D., Assistant in Zoology, Columbia University.
SCOTT, SISTER FLORENCE M., Assistant Professor of Biology, Seton Hill College.
SHAPIRO, HERBERT, Assistant in Zoology, Columbia University.
SHAW, C. RUTH, Graduate Assistant, University of Pittsburgh.
SKINNER, B. F., Graduate Student, Harvard University.
SMITH, SUZANNE G., Graduate Assistant in Zoology, University of Missouri.
SMYTHE, C. V., Fellow National Research Council, Rockefeller Institute.
SONNEBORN, TRACY M., Research Assistant, Johns Hopkins University.
STABLER, ROBERT M., Assistant Instructor, University of Pennsylvania.
STANCATI, MILTON F., Graduate Assistant, University of Pittsburgh.
STEWART, DOROTHY R., Assistant Professor of Biology, Skidmore College.
24 MARINE BIOLOGICAL LABORATORY
STEINER, MATTHEW M., Assistant in Biology, New York University.
STREET, SIBYL F., Assistant to Department of Zoology, Yassar College.
STUCK, FLORENCE, Student, Columbia University.
TANG, PEI-SUNG, Johns Hopkins University.
TE\YINKEL, Lois E., Assistant in Zoology, Barnard College, Columbia University.
TOOTH ILL, MARTHA C.. Assistant in Biology, Brown University.
TUAN, Hsu-CnvAN. Graduate Student. University of Pennsylvania.
TURNER, CLARENCE D., Research Assistant, University of Missouri.
WATERS, NELSON F., Research Assistant in Applied Botany. Harvard University.
WELLS, EVELYN, Instructor in Biology, St. Mary's Seminary.
WELLS, L. J.. Graduate Student, University of Chicago.
WILDE, MARY H., Graduate Assistant. New Jersey College for Women.
WRIGHT, CHARLES I., Fellow in Physiology, University of Rochester Medical
School.
YANCEY, PATRICK H., Graduate Student, St. Louis University.
Research Assistants
BECK, L. V., Teaching Fellow. New York University.
BERNSTEIN, ALAN. Research Assistant, New York University.
BUCK, LOUISE H., Research Assistant, New York University.
CAMPBELL, RAYMOND W.. Assistant. Fatigue Laboratory, Harvard University.
DAVIDSON, SYDNEY A., Williams College.
EISENBRANDT. W. II., Student, University of Maryland Medical School.
FRANCIS, DOROTHY S., Research Assistant, Memorial Hospital.
FRIEDMAN, HILDA, Assistant in Pathology. Washington University Medica! School.
FRIEDHEIM, ERNST A. H., Rockefeller Institute.
GRAND, CONSTANTINE G., Research Assistant, New York University.
GREENBERG, JACOB, Medical Student, Yale University Medical School.
HARRYMAN, ILENE, Research Assistant. Lilly Research Laboratory.
HEUSNF.K. A. P.. Student, Swarthmore College.
HOFFMAN, OLIVE D.. Research Assistant, New York University.
LAZAROVICH-HKEBELIANOVICH, MARA DE, Research Assistant, New York University.
MENDELSON, E. S.. Research Assistant, I'liivi-r^ity of Pennsylvania.
MENKIN, MIRIAM F.. Henry Phipps Institute.
OBERG, S. ALBERT, Harvard University.
PARKS, ELIZABETH K., Graduate Assistant in Zoology, Oberlin College.
RAVENSWAAY, VAN A. C.. Research Assistant, Washington University.
REYNOLDS, SARA JANE, Research Assistant, New York University.
RUDNICK, DOROTHEA. Research Assistant, University of Chicago.
SALOMON, KURT, Fellow of the Rockefeller Foundation. Rockefeller Institute for
Medical Research.
SANDERS, ROSALTHA, Technician, Yale University.
SCHUBERT, MAXWELL, Assistant, Rockefeller Institute for Medical Research.
SCOTT, ALLAN C., Graduate Assistant. University of Pittsburgh.
SIIATTUCK, G. EDGAR. Assistant Instructor in Physiology, University of Pennsyl-
vania.
SHEAR, M. J., Administrative Officer and Research Chemist, Pediatric Research
Laboratory. Jewish Hospital.
SHLAER, SIMON, Research Assistant, Columbia University.
SMITH, M. DOREEN, Research Assistant, Memorial Hospital.
SVVANN, SHERLOCK, Research Associate. University of Illinois.
TOCKER, ALBERT M.. Student, Washington University Medical School.
WALLACE, EDITH M.. Artist and Research Assistant, Carnegie Institution of Wash-
ington.
RKPORT OF THE DIRECTOR
Students
BOTANY
BREED, HKLKN L., Student. Wellesley Collar.
BRUNEL, JULES, Assistant Professor of Botany, University of Montreal.
CHEEVER, CLARENCE A., Boston Society of Natural History.
DROUET, FRANCIS, Graduate Assistant, University of Missouri.
FORBES, JOHN M.. Student, Harvard University.
GLIDDEN, DOROTHY P.. Student, Smith College.
HOPKINS, MILTON, Student, Amherst College.
HUNTINGTON, EVELYN, Student, Vassar College.
KLOSE, THEODORE G., Instructor in Botany, Yassar College.
LOUGHRIDGE, GASPER A., Laboratory Assistant in Botany. Rutgers University.
McKEE. JEWEL C., University of Wisconsin.
OPPENHEIMER, JANE M., Student, Bryn Mawr College.
ROLAND, ALBERT E., Student, Acadia University.
SAFFORD, DEC i us W., Dartmouth College.
STEWART, PAUL A., Student, University of Rochester.
WILDE. MARY H.. Assistant, New Jersey College for Women.
EMBRYOLOGY
BAKER, E. G. STANLEY. Student, DePauw University.
BALLARD, OVERTON T.. University of Illinois.
BAMBER, LYLE E., Graduate Student and Assistant. University of Illinois.
CARTER, GEORGE H., Student, Amherst College.
DAWSON, HELEN L., Graduate Student, Washington University.
DERBYSHIRE, ARTHUR J., JR.. Harvard University.
EARL, RUTH R., Technician. College of the City of Xew York.
EATON, THEODORE H., JR.. Cornell University.
GREEN, DAVID E., Assistant, Washington Square College.
GUERLAC, HENRY E., Student. Cornell University.
HILEMAN. CLARA M., Instructor in Biology, Penn State College.
HJORTLAND, ARTHUR L.. Assistant, University of Illinois.
HUNNINEN, ARNE V., Student, Wesleyan University.
JOHNSON, ARLENE C., Student, Wheaton College.
JOHNSON, MYRA L., Student, Smith College.
LANE, MARY, Smith College.
LOEFER, JOHN B., Graduate Assistant in Biology, New York University.
MACKMULL, GULDEN, Demonstrator of Histology and Embryology, Baugh Insti-
tute of Anatomy, Jefferson Medical College.
MAXWELL, FLORENCE J., Instructor, Carnegie Institute of Technology.
NICHOLS, ROWENA. Wellesley College.
PATCH, ESTHER M., Teacher of Biology, East Boston High School.
REID, MARION A., Instructor, Boston University Medical School.
RILEY, LENA C., Student, Wellesley College.
ROSENBAUM, LOUISE, Student, University of Pennsylvania.
SCOTT, JOHN P., University of Wyoming.
STANLEY. WILLARD F., Graduate Student and Research Assistant, University of
Illinois.
WELLS, LEMEN J., Graduate Student, University of Chicago.
PHYSIOLOGY
APPELROT, SAMUEL, Fellow, Rockefeller Foundation.
BEHNER, DOROTHY M., Assistant, New York University.
CHANG, TSUNG H., Graduate Student, University of Chicag •.
26 MARINE BIOLOGICAL LABORATORY
CHEN, TCANG T.. Assistant in Biochemistry, Poking Union Medical College.
DANN, MARGARET, Assistant in Physiology, Cornell University Medical College.
DREW, WILLIAM B., Massachusetts Agricultural College.
DuBois, DELAFIELD, New York University.
FENG, TE-PEI, Graduate Student, University of Chicago.
GARDNER, EDITH McN., Assistant in Physiology, Vassar College.
GATES, FREDERICK L., Research Fellow, Harvard University.
HEGNAUER, ALBERT II., Fellow in Physiology, University of Rochester Medical
School.
LEITCH, JAMES L., University of California.
MONAGHAN, BETTY R., Assistant and Graduate Student, Washington University.
OLIPHANT, JOSEPH F.. Instructor in Biology, Union College.
OSTER, ROBERT H., Student, Williams College.
PITTS, ROBERT F., Johns Hopkins University.
SHAW, GRETCHEN, Graduate Student, University of Chicago.
STEINER, MATTHEW M., Assistant in Biology, New York University.
STEVENS, TIIELMA O., Graduate Assistant, Mt. Holyoke College.
TANG, PEI-SUNG, Harvard University.
VACK, CHRISTINE M., Technician, Harvard University Medical School.
VICARI, EMILIA M.. Research Associate, Cornell University Medical College.
WOODWARD, ALVALYN E., Assistant Professor, University of Michigan.
PROTOZOOLOGY
BREHME, KATHERINE S., Barnard Colh
CARTER, HELEN D.. Elmira College.
COSTELLO, DoN.M.ii I'.. .v.iJ I Indrllu-rg. Detmit, Michigan.
EMBICH, JOHN R., Graduate Student, Columbia University.
FLAMMON, SISTER M. Mrun i., Sctmi Hill College.
FRYE, MARY ELIZABETH, Pennsylvania College fur Wnmen.
LHERISSON, CAMILLE. Professor of Biology, University of Haiti Medical School.
MASTEN. Lois I-'.., Khniru College.
MORGAN, \\'II.I.IK A.. Assistant Instructor in Biology, Coker College.
SCHOELT, ABRAHAM 11.. Graduate Student. Columbia University.
SCOTLAND, MINNIE B., Teacher, New York State College for Teachers.
SMITH, CLAIKI M., Hunter ColK-.ne.
STEINBERG, BERNHARD, Director of Laboratories and Research, Toledo Hospital.
WEISMAN, MAXWELL N., Fellow, College of the City of New York.
I \YKKTK!', RATE ZOOLOGY
ALDERMAN, EVANGELINE, Oberlin College.
BAUMGARTNKR, FREDERICK M.. Butler University.
BITTINGER, ISABEL, Radcliffe College.
BoARDMAN, EDWARD 'I'., Graduate Assistant, Johns Hopkins University.
BROWN, FRANK A., JR.. Harvard University.
CAMPBELL. DAX H., Student, Wahash College.
CARLSON, J. GORDON, Assistant in Zoology, University of Pennsylvania.
CHADWICK, CLAUDE S., Instructor in Biology, Vanderbilt University.
COHEN, ROSE S., Graduate Assistant, University of Cincinnati.
COLEMAN, LUCILLE, Agnes Scott College.
COPLAN, HELEN M., Student, Goucher College.
CRAIG. ROBERT L.. Student, Amherst College.
CROWELL. PRINCE S.. JR., Bowdoin College.
DEE, M. BARBARA, Assistant in Science, Jamaica Plain High School.
Di-:Roo. GRACE, Radcliffe College.
DERRICKSON. MARY B., Goucher College.
REPORT OF THE DIRECTOR
DORRIS, FRANCES S., Graduate Student, Yale University.
DOYLE, WILLIAM L., Johns Hopkins University.
EICHOLD, EVA C., Student, Newcomb College.
EVERETT, JOHN \V., Yale University.
FARBER, SEYMOUR M., University of Buffalo.
FENNELL, R. A., Graduate Student, Duke University.
FISHER, KENNETH C., Student and Assistant, Acadia University.
FRENCH, CHARLES S., Harvard University.
HAMBURGER, Louis P., JR., Johns Hopkins University.
HART, HELEN B., Student, Wellesley College.
HASTINGS, MARGARET, Student, Mt. Holyoke College.
HAYES, FREDERICK R., Assistant Professor of Zoology, Dalhousie University.
HEISS, ELIZABETH M., Assistant in Biology and Histology, Purdue University.
HEUSNER, ALBERT P., Swarthmore College.
HEWITT, DOROTHY C., Graduate Student, Yale University.
HOLLOWAY, MAY P., Teacher of Science, Burke School.
HUBBARD, RUTH A., Assistant, Cleveland Museum of Natural History.
JACKSON, JEANNETTE A., Graduate Assistant in Zoology, Syracuse University.
JOHNSON, DOROTHY F., Laboratory Assistant, Wellesley College.
KILLE, FRANK R., Assistant Professor of Biology, Birmingham-Southern College.
KROC, ROBERT L., Graduate Assistant, University of Wisconsin.
LEAVITT, BENJAMIN B., Instructor in Zoology, Dartmouth College.
MENZEL. ARTHUR E. O., National Tuberculosis Research Fellow, Presbyterian
Hospital.
MERRIMAN, DANIEL, Student, Harvard University.
MORRIS, SAMUEL, Instructor in Zoology, University of Pennsylvania.
PREST, MARGARET R., Graduate Assistant, Mt. Holyoke College.
REDMOND, ALBERT C., Student, Hamilton College.
REYNOLDS, ALBERT E., Assistant in Zoology, DePauw University.
RITTER, M. ESTHER, Student, Wilson College.
SHEA, MARGARET, Student, Oberlin College.
SIDEBOTHAM, RUTH S., Graduate Assistant in Zoology, Washington University.
SNELL, PETER A., Fellow in Biology, Princeton University.
SNOOK, THEODORE, Graduate Assistant in Zoology, Rutgers University.
SWANSON, OSCAR E., Student, Antioch College.
TIPTON, SAMUEL R., Graduate Student, Duke University.
TOWNSEND, GRACE, Instructor, Joliet Township High School and Junior College.
TREAT, DOROTHY A., Assistant in Department of Education, Cleveland Museum of
Natural History.
WATERS, NELSON F., Graduate Student, Harvard University.
WEED, MILTON R., Student, Wesleyan University.
WOODRUFF, BETH H.. Graduate Assistant, Western Reserve University.
28
MARINE BIOLOGICAL LABORATORY
TABULAR VIEW OF ATTENi ) AXCE
1926 1927 1928 1929 1930
INVESTIGATORS— Total 252 294 323 329 337
Independent 156 209 217 234 217
Under Instruction 84 57 81 71 87
Research Assistants 12 28 25 24 33
STUDENTS— Total 141 141 133 125 136
Zoology 56 57 57 53 56
Protozoology 19 17 16 15 14
Embryology 28 32 29 28 27
Physiology 18 19 15 17 23
Botany 20 16 16 12 16
TOTAL ATTENDANCE 393 435 456 454 473
Less Persons registered as both students
and investigators 8 1 2 10 14
385 434 454 444 459
INSTITUTIONS HI-PHI si- NT-HI)— Total 119 111 111 123 126
By Investig-;it. .!> 84 8'» 80 96 95
By Students 60 (,3 66 64 71
SCHOOLS AND ACADEMIES RH.I'RESKNTED
My Investigators 1 1
My Students 4 4 1 1 4
r.iKF.icN INSTITUTIONS RKI-RKSKNTI.H
By Invest igator> 17 15 13 30 7
Bv Student- . 38832
4. SUBSCRIBING AXD COOPERATING INSTITUTIONS
Acadia University
. \mherst College
Antioch College
l'.ea\ er < 'ollege
Bowdoin College
Brown University
I'.rvn Mawr College
Rutler College
C. R. B. F.ducational Foundation
California Institute of Technology
Carnegie Institution of Washington
Chinese Educational Mission
Columbia University
Cornell University
Cornell University Medical College
Dalhousie University
Dartmouth College
I Vl'amv University
Duke University
F.lmira College
General Education Board
Goucher College
Hamilton College
I harvard University
Harvard University Medical School
Howard University
I lunter College
Indiana University
Industrial & Engineering Chemistry.
of the American Chemical Society
Johns Hopkins University
Johns Hopkins University Medical
Scliool
F.li Lilly & Co.
Memorial Hospital of X. Y. City
rUorehouse College
Mount TTolyoke College
National Research Council
New York State Department of
Health
New York University
Oherlin College
REPORT OF THE DIRECTOR
29
Pennsylvania College for Women
Princeton University
Radcliffe College
Rockefeller Foundation
Rockefeller Institute for Medical
Research
Rutgers University
Seton Hill College
Smith College
Sophie Newcomb College
Southwestern
St. Louis University
Swarthmore College
Tufts College
Union College
United States Dept. of
University of Buffalo
University of Chicago
University of Cincinnati
University of Illinois
University of Michigan
University of Missouri
University of Nebraska
Universitv of Pennsylvania
Agriculture
University of Pennsylvania Medical
School
University of Pittsburgh
University of Rochester
University of Tennessee
University of Virginia
University of Wisconsin
Vanderbilt University Medica
School
Yassar College
Wabash College
Washington University
Washington University Medical
School
AVellesley College
Wesleyan University
Western Reserve University
West Virginia State College
West Virginia University
Wheaton College
Wistar Institute of Anatomy and
Biology
Yale Universitv
SCHOLARSHIP TABLES
Lucretia Crocker Scholarships for Teachers in Boston.
Scholarship of $100 supported by a friend of the Laboratory since 1898.
The Edwin S. Linton Memorial Scholarship of Washington and Jefferson
College.
The Bio Club Scholarship of the College of the City of New York.
Ida H. Hyde Scholarship of the University of Kansas.
5. EVENING LECTURES, 1930
Tuesday, July 1
DR. W. B. SCOTT " Xew Light on the Development and
Migrations of American Mam-
mal.-."
"Prospects and Problems of Ocean-
ography."
Tuesday, July 8
DR. HENRY B. BIGELOW
Tuesday, July 15
DR. H. S. IENNINGS .
"Heredity and Mutation in Relation
to Environment in Protozoa."
Friday, July 18
DR. EDWARD MELLAXBY " Food Deficiencies."
Tuesday. July 22
DR. F. R. LILLIE " The Action of the Sex Hormones in
the Fowl : An Account of the Chi-
cago Experiments."
30 MARINE BIOLOGICAL LABORATORY
Tuesday, July 29
DR. LEONOR MICHAELIS " The Reversible Oxidizable-reducible
Systems Found in Living Organ-
isms."
Tuesday, August 5
DR. E. F. DuBois " Recent Progress in the Field of Re-
spiratory Metabolism."
Tuesday, August 12
DR. E. CHATTON "A Study of the Dinorlagellate, Poly-
krikos Schwartzi as a Basis for the
Discussion of Some Problems of
General Cytology."
Tuesday, August 19
Tin-: REYNOLD A. SPAETH Mi M-
RIAL LECTURE, delivered by DR.
HARDOLPH WASTEXEYS " Protein Synthesis."
Thursday, August 21
DR. II. H. GRAN " The Productivity of the Ocean."
Tuesday, August 26
DR. D. \Y. BKOXK " Xerve Impulse Rhythms and the
Control of Movement."
Tuesday, September 2
DR. G. C. HIRSCII " The Problem of Restitution with
Special Regard to the Phenomena of
Secretion."
Tuesday, September 9
DR. G. E. HUTCHINSON "The Hydrobiology of Arid and
Semi-arid Regions."
SPECIAL LECTURES AND MOTION PICTURES
Monday, August 1 1
" The lllyria Expedition to the Galapagos, the South Sea Islands, the Xew
Hebrides, the Solomon Islands, Xew Guinea, Bali and Angkor." Exhibited
by MR. CORNII irs CRANE, the Leader of the Expedition.
Thursday, August 14
" The Florida Everglades and the Proposed Tropic Everglades National
Park." Illustrated with colored lantern slides. MR. ERNEST F. COE,
Chairman of the Tropic Everglades Park Association.
Thursday, August 28
Motion pictures.
"William Harvey and the Circulation of the Blood." Arranged by SIR
THOMAS LEWIS.
"The Early Development of the Rabbit Egg." Du. \Y. II. LEWIS and DR.
I'. W. GREGORY.
"The Life Cycle of the Oyster." Prepared by the AMERICAN MUSEUM
OK NATURAL HISTORY.
Friday, August 29
" Motion Pictures of Marine and Fresh Water Protozoa of the Woods
Hole Region," RTTH B. I IOWI.AXD, Department of Biology, Xew York
University.
REPORT OF THE DIRECTOR
6. SHORTER SCIENTIFIC PAPERS, 1930
Tuesday, June 24
DR. BALDUIN LUCKE AND
DR. MORTON McCuTCHEON "The Effect of Injury on Cellular
Permeability to Water."
DR. M. M. BROOKS "The Relation between rH and the
Penetration of Oxidation-reduction
Indicators."
DR. S. C. BROOKS " Accumulation of Ions in Living
Cells."
DR. M. H. JACOBS, iMR. A. K. PAR-
PART, DR. W. A. SMITH AND MR.
G. E. SHATTUCK " The Permeability of the Erythrocyte
to Urea."
Friday, June 27
DR. B. H. WILLIER " The Developmental Relations of the
Heart and the Liver in Chorio-
allantoic Grafts."
DR. E. E. JUST " Cortical Protoplasm and Vital Phe-
nomena."
DR. WILLIAM C. YOUNG " The Post-testicular History of Sper-
matozoa and Reproduction in the
Male Guinea Pig."
DR. L. V. HEILBRUNN " The Action of Ultra-violet Rays on
Arbacia Egg Protoplasm."
Thursday, July 3
DR. WALTER S. ROOT AND
DR. CHARLOTTE HAYWOOD " The Effect of Carbon Dioxide upon
the Rate of Oxygen Consumption
and of Cleavage of the Arbacia
Egg."
DR. L. MlCHAELIS AND
DR. K. SALOMON " Respiration of Ervthrocytes."
DR. R. W. GERARD " Observations on the Metabolism of
the Coccus, Sarcina lutea."
DR. E. S. G. BARRON " The Effect of Methylene Blue upon
the Respiration of Normal and of
Cancer Tissue."
Friday, July 11
DR. W. R. TAYLOR " Chromosome Structure in Meiosis of
Gasteria."
DR. B. M. DUGGAR " New Technique and Some Adsorp-
tion Studies with Virus Diseases of
Plants."
Thursday, July 17
DR. VERA KOEHRIXG "Some Cytological Relationships in
Narcosi-."
DR. ELEANOR H. SLIFER " The Mitotic Activity in the Devel-
oping Grasshopper Egg."
MARINE BIOLOGICAL LABORATORY
DR. A. F. HUETTXER " Spermatogenesis in Drosophila mel-
anogaster."
DR. HOPE HiBBARn " Cytological Studies on the Silk Gland
of Bombyx mori."
Friday, July 25
DR. S. GELFAX " The All-or-Xone Law in Muscle."
DR. \Y. R. AMBERSOX. MR. A. K.
PARPART AXD Miss GERTRUDE SAN-
DERS " Low Voltage Elements of the Action
Potential \\'ave of Nerve."
DR. F. O. SCHMITT " The Effect of Cyanides and Carbon
Monoxide on Nerve."
Saturday, July 26
DR. M. Driirissox "Cardiac Automatism in Inverte-
brates."
DR. \V. F. GAKREV " Observations on the 1 leart of Limu-
lus."
DR. D. J. EDWARDS " The Action of Pressure on Some
1 'hysiological Processes."
DR. G. H. BISHOP " The Influence of lodo-acetic Acid on
Muscle Contracture."
DR. F. H. PRATT " Experiments on the Terminal Nerve-
muscle Unit."
DR. I). \V. BROXK " Graded Mu>cular Contractions."
DR. K. \\'. GERARD " Nerve Metabolism and Asphyxia."
DR. R. S. LII.I.IE " Recovery in the Passive Iron Wire
Model.""
DR. A. M. MOXXIKR " Mathematical Analysis Applied to
the Functions of the Nervous Sys-
tem."
DR. 1 1. DAVIS " Re-education and Modification of
Reflexes."
DR. G. P. McCorrii " Patterns of Some Extra-ocular Re-
tle\e> in the Cat."
DR. P. I! \K-i) " The Piehavior of a Cat without the
Telencephalon."
Friday, Au.uu-t 1
DR. PAUL S. llr.\sir.\w "Some Biological Effects Produced
by Alpha Particles on Drosophila
Eggs."
MR. WARE CATTEI.I " The Effect of X-Rays upon the
Growth of the Wheat Seedling."
DR. W. C. CURTIS " Effects of X-Rays upon Regenera-
tion."
DR. CHARLES PACKARD "The Relation between Division Rate
and Susceptibility to Radiation."
Friday, August 8
DR. HELEN M. MILLER "Life Cycle of a Bisexual Rotifer."
DR. TRACY M. SONNEBORX "Cause, Inheritance, and Effects of
the Chain-forming Tendency in the
Ciliate Protozoan, Colpidium."
REPORT OF THE DIRECTOR
DR. RUTH STOCKING LYNCH " The Effects of Long-continued
Starvation in a Rotifer in Relation
to Inheritance."
DR. H. W. STUNKARD " The Life History of Cryptocotyle
lingua."
Friday, August 15
DR. MARY S. M \cDoucALL "A Mutation in Chilodon uncinatus
Produced by Ultra-violet Radiation
—A Preliminary Report."
DR. R. D. MANWELL " The Effect of Quinine and Plas-
moquin on Avian Malaria."
DR. E. CHATTON " The Asymmetric Motile Stages of
Epistylis and the Question of the
So-called Longitudinal Division of
the Vorticellidae."
DR. RUTH B. ROWLAND " Cine-photomicrograph of Microin-
jection of Vacuolated Protoplasm."
Friday, August 22
DR. CALVIN B. BRIDGES " The Neutralization of the Effects of
Deficiencies through Duplications of
the Same Chromosome Material."
DR. A. H. STURTEVANT "A Peculiar Sex-ratio in Drosophila
obscura."
DR. HELEN REDFIELD " Studies of Crossing-over in Droso-
phila."
DR. JACK SCHULTZ " The Eye of Pigments of Drosophila."
Friday, August 29
DR. H. B. GOODRICH AND
MR. I. B. HANSEN "Embryonic Development of Men-
del ian Characters in the Goldfish."
DR. H. A. ABRAMSON " The Isoelectric Point of Mammalian
Red Blood Cells."
DR. E. N. HARVEY AND
MR. A. L. LOOMIS " The Microscope-centrifuge."
Friday, September 5
DR. J. M. JOHLIN " The Extraction of Micro-organisms."
DR. W. E. GARREY AND
DR. W. R. BRYAN " Alkalosis in Relation to Tetany fol-
lowing High Temperatures after
Parathyroidectomy."
DR. K. BLANCHARD " Catalysis of Condensation Reactions
by Amino-acids."
DR. L. MlCHAELIS AND
DR. M. SCHUBERT " Metal Complex Compounds of Thio-
glycollic Acid,"
3
34 MARINE BIOLOGICAL LABORATORY
7. MEMBERS OF THE CORPORATION
1. LIFE MEMBERS
ALLIS, MR. E. P., JR., Palais Carnoles, Menton, France.
ANDRF.WS, MRS. GWENDOLEN FOULKE, Baltimore, Md.
BILLINGS, MR. R. C., 66 Franklin St., Boston, Mass.
CONKLIN, PROF. EDWIN G., Princeton University, Princeton, N. J.
COOLIDGE, MR. C. A., Ames Building, Boston, Mass.
CRANE, MR. C. R., New York City.
EVANS, MRS. GLENDOWER, 12 Otis Place, Boston, Mass.
FAY, Miss S. B., 88 Mt. Vernon St., Boston, Mass.
FOOT, Miss KATHERIXK, Care of Morgan Harjes Cie, Paris, France.
GARDINER, MRS. E. G., Woods Hole, Mass.
JACKSON, Miss M. C., 88 Marlboro St., Boston, Mass.
JACKSON, MR. CHAS. C., 24 Congress St., Boston, Mass.
KIDDER, MR. NATHANIEL T., Milton, Mass.
KING, MR. CHAS. A.
LEE, MRS. FREDERIC S., 279 Madison Are., New York City, N. Y.
LOWELL, MR. A. LAWRENCE, 17 Quincy St., Cambridge, Mass.
MEANS, DR. JAMES HOWARD, 15 Chestnut St., Boston. Mass.
MERRIMAN, MRS. DANIEL, 73 Bay State Road, Boston, Mass.
MINNS, Miss SUSAN, 14 Louisburg Square, Boston, Mass.
MORGAN, MR. J. PIERPONT, JR., Wall and Broad Sts., New York City,
N. Y.
MORGAN, PROF. T. H., Director of Biological Laboratory, California
Institute of Technology, Pasadena, Calif.
MORGAN, MRS. T. H., Pasadena, Calif.
NOYES, Miss EVA J.
OSBORN, PROF. HENRY F., American Museum of Natural History. New
York, N. Y.
PHILLIPS, MRS. JOHN C., Windy Knob, Wenham, Mass.
PORTER, DR. H. C., University of Pennsylvania, Philadelphia, Pa.
SEARS, DR. HKXKY F., Xf> Bracon St., Boston, Mass.
SHEDD, MR. E. A.
THORNDIKE, DR. EDWARD L., Teachers College, Columbia University,
New York City, N. Y.
TRELEASE, PROF. WILLIAM, University of Illinois, Urbana, 111.
WARE, Miss MARY L., 41 Brimmer St., Boston, Mass.
WILLIAMS, MRS. ANNA P., 505 Beacon St., Boston, Mass.
WILSON, DR. E. B., Columbia University, New York City, N. Y.
REPORT OF THE DIRECTOR
2. REGULAR MEMBERS, AUGUST, 1930
ADAMS, DR. A. ELIZABETH, Mount Holyoke College, South Hadley,
Mass.
ADDISON, DR. W. H. F., University of Pennsylvania Medical School,
Philadelphia, Pa.
ADOLPH, DR. EDWARD F., University of Rochester, School of Medicine
and Dentistry, Rochester, N. Y.
ALLEE, DR. W. C., University of Chicago, Chicago, 111.
ALLEN, PROF. CHARLES E., University of Wisconsin, Madison, Wis.
ALLEN, PROF. EZRA, New York Homeopathic Medical College, New
York City, N. Y.
ALLYN, DR. HARRIET M., Mount Holyoke College, South Hadley, Mass.
AMBERSON, DR. WILLIAM R., University of Tennessee, Memphis, Tenn.
ANDERSON, DR. E. G., California Institute of Technology, Pasadena,
Calif.
AUSTIN, DR. MARY L., Wellesley College, Wellesley, Mass.
BAITSELL, DR. GEORGE A., Yale University, New Haven, Conn.
BAKER, DR. E. H., 5312 Hyde Park Boulevard, Hyde Park Station,
Chicago, 111.
BALDWIN, DR. F. M., University of Southern California, Los Angeles,
Calif.
BECKWITH, DR. CORA J., Vassar College, Poughkeepsie, N. Y.
BEHRE, DR. ELINOR H., Louisiana State University, Baton Rouge, La.
BENNITT, DR. RUDOLF, University of Missouri, Columbia, Mo.
BIGELOW, PROF. R. P., Massachusetts Institute of Technology, Cam-
bridge, Mass.
BINFORD, PROF. RAYMOND, Guilford College, Guilford College, N. C.
BISSONNETTE, DR. T. H., Trinity College, Hartford, Conn.
BLANCHARD, PROF. KENNETH C., New York University, Washington
Square College, New York City, N. Y.
BODINE, DR. J. H., University of Iowa, Iowa City, la.
BORING, DR. ALICE M., Yenching University, Peking, China.
BOWLING, Miss RACHEL, Columbia University, New York City, N. Y.
Box, Miss CORA M., University of Cincinnati, Cincinnati, O.
BRADLEY, PROF. HAROLD C., University of Wisconsin, Madison, Wis.
BRAILEY, Miss MIRIAM E., 800 Broadway, Baltimore, Md.
BRIDGES, DR. CALVIN B., California Institute of Technology, Pasadena,
Calif.
BRONK, DR. D. W., University of Pennsylvania, Philadelphia. Pa.
BROOKS, DR. S. C., University of California, Berkeley, Calif.
BUCKINGHAM, Miss EDITH N., Sudbury, Mass.
36 MARINE BIOLOGICAL LABORATORY
BUDINGTON, PROF. R. A., Oberlin College, Oberlin, O.
BULLINGTON, DR. W. E., Kaiiclolph-Macon College, Ashland, Ya.
BUMPUS, PROF. H. C., 76 Carlton Road. \\'al)an, Mass.
BYRNES, DR. ESTHER F., 1803 North Camac Street, Philadelphia. Pa.
CALKINS, PROF. GARY X., Columbia University, New York City, N. Y.
CALVERT, PROF. PHILIP P., University of Pennsylvania, Philadelphia,
Pa.
CARLSON, PROF. A. J., University of Chicago, Chicago, 111.
CAROTHERS, DR. ELEANOR E., University of Pennsylvania, Philadelphia.
Pa.
CARROLL, PROF. MITCHEL, Franklin and Marshall College, Lancaster,
Pa.
CARVER, PROF. GAIL L., Mercer University, Macon, Ga.
CATTELL, DR. McKEEN, Cornell University Medical College, New
York City, N. Y.
CATTELL, PROF. J. McKEEN, Garrison-on-Hudson, N. Y.
CATTELL, MR. WARE, Garrison-on-Hudson, N. Y.
CHAMBERS, DR. ROBERT, Washington Square College, New York Uni-
versity, Washington Square, New York City, N. Y.
CHARLTON. DR. HARRY H., University of Missouri Columbia, Mo.
CHATTON, DR. EDOUARD. University of Strasbourg, Strasbourg, France.
CIIIDESTER, PROF. F. E., West Virginia University, Morgantown,
W. Va
CHILD, PROF. C. M., University of Chicago, Chicago, 111.
CLAPP, PROF. CORNELIA M., Montague, Mass.
CLARK, PROF. E. R., University of Pennsylvania, Philadelphia, Pa.
CLELAND, PROF. RALPH E., Gouchcr College, Baltimore, Md.
CLOWES, PROF. G. H. A., Eli Lilly & Co., Indianapolis, Ind.
COE, PROF. W. R., Yale University, New Haven, Conn.
COHN, DR. EDWIN J., 183 Brattle St., Cambridge, Mass.
COLE, DR. I ALBERT C., Williams College, Williamstown, Mass.
COLE, DR. LEON J., College of Agriculture, Madison. \YU.
COLLETT, DR. MARY E., Western Reserve University, Cleveland, O.
COLLEY, MRS. MARY W., 36 Argyle Place, Rockville Centre, Long
Island, N. Y.
COLTON, PROF. II. S., Box 127, Flagstaff, Ariz.
CONNOLLY, DR. C. J., Catholic University, Washington, D. C.
COPELAND, PROF. MANTON, Bowdoin College, Brunswick. Me.
COWDRY, DR. E. V., Washington University, St. Louis, Mo.
CRAMPTON, PROF. II. E., Barnard College, Columbia University, New
York City, N. Y.
CRANE, MRS. C. R., Woods Hole, Mass.
RFPORT OF THE DIRECTOR
CURTIS, DR. MAYNIK R., Crocker Laboratory, Columbia University,
New York City, N. Y.
CURTIS. PROF. W. C., University of Missouri, Columbia, Mo.
DAVIS, DR. ALICE R., Castle Point. Hoboken, N. J.
DAVIS, DR. DONALD W., College of William and Mary, Williamsburg,
Va.
DAWSON, DR. A. B., Harvard University, Cambridge, Mass.
DAWSON, DR. J. A., The College of the City of New York, New York
City, N. Y.
DEDERER, DR. PAULINE H., Connecticut College, New London, Conn.
DELLINGER, DR. S. C., University of Arkansas, Fayetteville, Ark.
DODDS, PROF. G. S., Medical School, University of West Virginia, Mor-
gantown, W. Va.
DOLLEY, PROF. WILLIAM L., University of Buffalo, Buffalo, N. Y.
DONALDSON, PROF. H. H., Wistar Institute of Anatomy and Biology,
Philadelphia, Pa.
DONALDSON, DR. JOHN C., University of Pittsburgh, School of Med-
icine, Pittsburgh, Pa.
DREW, PROF. OILMAN A., Eagle Lake, Florida.
Du Bois, DR. EUGENE F., Cornell University Medical College, New
York City, N. Y.
DUGGAR, DR. BENJAMIN M., University of Wisconsin, Madison, Wis.
DUNGAY, DR. NEIL S., Carleton College, Northfield, Minn.
DUNN, DR. L. C., Columbia University, New York City, N. Y.
EDWARDS, DR. D. J., Cornell University Medical College, New York
City, N. Y.
ELLIS, DR. F. W., Monson, Massachusetts.
FARNUM, DR. LOUISE W., Hsiang-Ya Hospital, Changsha, Hunan,
China.
FAURE-FREMIET, PROF. EMMANUEL, College de France, Paris, France.
FENN, DR. W. O., Rochester University, School of Medicine, Rochester,
N. Y.
FIELD, Miss HAZEL E., Occidental College, Los Angeles, Calif.
FORBES, DR. ALEXANDER, Harvard University Medical School, Boston,
Mass.
FRY, DR. HENRY J., Washington Square College, New York City, N. Y.
GAGE, PROF. S. H., Cornell University, Ithaca, New York.
GARREY, PROF. W. E., Vanderbilt University Medical School, Nashville.
Term.
GATES, DR. F. L., 31 Fayerweather St., Cambridge, Mass.
GATES, PROF. R. RUGGLES, University of London, London, England.
GEISER, DR. S. W., Southern Methodist University, Dallas, Tex.
38 MARINE BIOLOGICAL LABORATORY
GLASER, PROF. O. C, Amherst College, Amherst, Mass.
GLASER, PROF. R. W., Rockefeller Institute for Medical Research,
Princeton, N. J.
GOLDFORB, PROF. A. J., College of the City of Xew York, Xcw York
City, N. Y.
GOODRICH, PROF. H. B., Wesleyan University, Middletown, Conn.
GRAHAM, DR. J. Y., University of Alabama, University, Ala.
GRAVE, PROF. B. H., DePauw University, Greencastle, Ind.
GRAVE, PROF. CASWELL, Washington University, St. Louis, Mo.
GRAY, PROF. IRVING E., Duke University, Durham, X. C.
GREENMAN, PROF. M. J., Wistar Institute of Anatomy and Biology,
Philadelphia, Pa.
GREGORY, DR. LOUISE H., Barnard College, Columbia University, New
York City, N. Y.
GUTHRIE, DR. MARY J., University of Missouri, Columbia, Mo.
GUYER, PROF. M. F., University of Wisconsin, Madison, Wis.
HAGUE, DR. FLORENCE, Sweet Briar College, Sweet Briar, Va.
HALL, PROF. FRANK G., Duke University, Durham, X1. C.
HANCE, DR. ROBERT T., University of Pittsburgh, Pittsburgh,. Pa.
HARGITT, PROF. GEORGE T., Duke University, Durham. X. C.
HARMAN, DR. MARY T., Kansas State Agricultural College, Manhattan,
Kans.
HARPER, PROF. R. A., Columbia University, New York City, N. Y.
HARRISON, PROF. Ross G., Yale University, New Haven, Conn.
HARVEY, MRS. E. N., Princeton, N. J.
HARVEY, PROF. E. N., Princeton University, Princeton, N. J.
HAYDEN, DR. MARGARET A., Wellesley College, Wellesley, Mass.
HAYWOOD, DR. CHARLOTTE, Mount llolyoke College, South Hadley,
Mass.
HAZEN, DR. T. E., Barnard College, Columbia University, New York
City, N. Y.
HEATH, PROF. HAROLD, Pacific Grove, California.
HECHT, DR. SELIG, Columbia University, New York City, XT. Y.
HEGNER, PROF. R. W., Johns Hopkins University, Baltimore, Md.
HEILBRUNN, DR. L. V., University of Pennsylvania. Philadelphia, Pa.
HESS, PROF. WALTER N., Hamilton College, Clinton, N. Y.
HINRICHS, DR. MARIE A., University of Chicago, Chicago, 111.
HISAW, DR. F. L., University of Wisconsin, Madison, Wis.
HOADLEY, DR. LEIGH, Harvard University, Cambridge, Mass.
HOGUE, DR. MARY J., 503 N. High St., West Chester, Pa.
HOLMES, PROF. S. J., University of California. Berkeley, Calif.
HOOKER, PROF. DAVENPORT, University of Pittsburgh, Pittsburgh, Pa.
REPORT OF THE DIRECTOR
39
HOPKINS, DR. HOYT S., New York University, College of Dentistry,
New York City, N. Y.
HOWARD, DR. HARVEY J., Washington University, St. Louis, Mo.
HOWE, DR. H. E., 2702 36th St., N. W., Washington, D. C.
HOYT, DR. WILLIAM D., Washington and Lee University, Lexington,
Va.
HUMPHREY, MR. R. R., University of Buffalo, School of Medicine,
Buffalo, N. Y.
HYMAN, DR. LIBBIE H., University of Chicago, Chicago, 111.
INMAN, PROF. ONDESS L., Antioch College, Yellow Springs, O.
IRWIN, DR. MARIAN, Rockefeller Institute, New York City, N. Y.
JACKSON, PROF. C. M., University of Minnesota, Minneapolis, Minn.
JACOBS, PROF. MERKEL H., University of Pennsylvania, Philadelphia,
Pa.
JENNINGS, PROF. H. S., Johns Hopkins University, Baltimore, Md.
JEWETT, PROF. J. R., Harvard University, Cambridge, Mass.
JOHNSON, PROF. GEORGE E., State Agricultural College, Manhattan,
Kans.
JONES, PROF. LYNDS, Oberlin College, Oberlin, O.
JUST, PROF. E. E., Howard University, Washington, D. C.
KEEFE, REV. ANSELM M., St. Norbert College, West Depere, Wis.
KENNEDY, DR. HARRIS, Readville, Mass.
KINDRED, DR. J. E., University of Virginia, Charlottesville, Va.
KING, DR. HELEN D., Wistar Institute of Anatomy and Biology, Phila-
delphia, Pa.
KING, DR. ROBERT L., State University of Iowa, Iowa City, la.
KINGSBURY, PROF. B. F., Cornell University, Ithaca, N. Y.
KIRKHAM, DR. W. B., Springfield College, Springfield. Mass.
KNAPKE, REV. BEDE, St. Bernard's College, St. Bernard, Ala.
KNOWER, PROF. H. McE., Albany Medical College, Albany, N. Y.
KNOWLTON, PROF. F. P., Syracuse University, Syracuse, N. Y.
KOSTIR, DR. W. J., Ohio State University, Columbus, O.
KRIBS, DR. HERBERT, 202A Copley Road, Upper Darby, Pa.
KUYK, DR. MARGARET P., Westbrook Ave., Richmond, Va.
LANCEFIELD, DR. D. E., Columbia University, New York City, N. Y.
LANGE, DR. MATHILDE M., Wheaton College, Norton, Mass.
LEE, PROF. F. S., College of Physicians and Surgeons, New York City,
N. Y.
LEWIS, PROF. I. F., University of Virginia, Charlottesville, Va.
LEWIS, PROF. W. H., Johns Hopkins University, Baltimore, Md.
LILLIE, PROF. FRANK R., University of Chicago, Chicago, 111.
LILLIE, PROF. RALPH S., University of Chicago, Chicago, 111.
40 MARINE BIOLOGICAL LABORATORY
LINTON, PROF. EDWIN, University of Pennsylvania, Philadelphia, Pa.
LOEB, PROF. LEO, Washington University Medical School, St. Louis,
.Mo.
LOEB, MRS. LEO, 812 Boland Place, St. Louis, Mo.
LOWTHER, MRS. FLORENCE DeL., Barnard College, Columbia University,
New York City, N. Y.
LUCKE, PROF. BALDUIN, University of Pennsylvania, Philadelphia. Pa.
LUND, DR. E. J., University of Texas, Austin, Tex.
LUSCOMBE, MR. W. O., Woods Hole, Mass.
LYNCH, DR. CLARA J., Rockefeller Institute, New York City, N. Y.
LYNCH, DR. RUTH STOCKING, Johns Hopkins University, Baltimore,
Md.
LYON, PROF. E. P., University of Minnesota, Minneapolis, Minn.
MACDOUGALL, DR. MARY S., Agnes Scott College, Decatur, Ga.
McCLUNG, PROF. C. E., University of Pennsylvania, Philadelphia, Pa.
McGEE, DR. ANITA NEWCOMB, Box 363, Southern Pines, N. C.
MCGREGOR, DR. J. H., Columbia University, New York City, N. Y.
McMuRRiCH, PROF. J. P., University of Toronto, Toronto, Canada.
McNAiR, DR. G. T., 1624 Alabama St., Lawrence, Kans.
MACKLIN, DR. CHARLES C., School of Medicine, University of Western
Ontario, London, Canada.
MALONE, PROF. E. F., University of Cincinnati, Cincinnati, O.
MANWELL, DR. REGINALD D., School of Hygiene and Public Health,
Johns Hopkins University, Baltimore, Md.
MARTIN, PROF. E. A., College of the City of New York, New York
City, N. Y.
MAST, PROF. S. O., Johns Hopkins University, Baltimore, Md.
MATHEWS, PROF. A. P., University of Cincinnati, Cincinnati, O.
MAYOR, PROF. JAM is \V., Union College, Schenectady, N. Y.
MEDES, DR. GRACE, University of Minnesota, Minneapolis, Minn.
MEIGS, DR. E. B., Dairy Division Experiment Station, Beltsville. Md.
MEIGS, MRS. E. B., 1736 M St., N. W., Washington, D. C.
METCALF, PKOF. M. M., Johns Hopkins University, Baltimore, Md.
METZ, PROF. CHARLES W., Carnegie Institution of Washington, Cold
Spring Harbor, Long Island, N. Y.
MICHAELIS, DR. LEONOR, Rockefeller Institute, New York City, N. Y.
MILLER, DR. HELEN M., Johns Hopkins University, Baltimore. Md.
MINER, DR. ROY W., American Museum of Natural History, New
York City, N. Y.
MITCHELL, DK. PHILIP II., Brown University, Providence. R. I.
MOORE, DR. CARL R., University of Chicago, Chicago, 111.
MOORE, PROF. GEORGE T., Missouri Botanical Garden. St. Louis, Mo.
REPORT OF THE DIRECTOR 41
MOORE, PROF. J. PI-:RCY, University of Pennsylvania, Philadelphia, Pa.
MORGULIS, DR. SERGIUS, University of Nebraska, Lincoln, Nebr.
MORRILL, PROF. A. D., Hamilton College, Clinton, N. Y.
MORRILL, PROF. C. V., Cornell University Medical College, New York
City, N. Y.
MULLER, DR. H. J., University of Texas, Austin, Tex.
NABOURS, DR. R. K., Kansas State Agricultural College, Manhattan,
Kans.
NEAL, PROF. H. V., Tufts College, Tufts College, Mass.
NEWMAN, PROF. H. H., University of Chicago, Chicago, 111.
NICHOLS, DR. M. LOUISE, Dreycott Apartments, Haverford, Pa.
NOBLE, DR. GLADWYN K., American Museum of Natural History, New
York City, N. Y.
NONIDEZ, DR. JOSE F., Cornell University Medical College, New York
City, N. Y.
OKKELBERG, DR. PETER, University of Michigan, Ann Arbor, Mich.
OSBURN, PROF. R. C., Ohio State University', Columbus, O.
OSTERHOUT, PROF. \Y. ]. V., Rockefeller Institute, New York City,
N. Y.
PACKARD, DR. CHARLES, Columbia University, Institute of Cancer Re-
search, 1145 Amsterdam Ave., New York City, N. Y.
PAGE, DR. IRVINE H., Presbyterian Hospital, New York City, N. Y.
PAPANICOLAOU, DR. GEORGE N., Cornell University Medical College,
New York City, N. Y.
PAPPENHEIMER, DR. A. M., Columbia University, New York City,
N. Y.
PARKER, PROF. G. H., Harvard University, Cambridge, Mass.
PATON, PROF. STEWART, Princeton University, Princeton, N. J.
PATTEN, DR. BRADLEY M., Western Reserve University, Cleveland, O.
PATTEN, PROF. WILLIAM, Dartmouth College, Hanover, N. H.
PAYNE, PROF. F., University of Indiana, Bloomington, Ind.
PEARL, PROF. RAYMOND, Institute for Biological Research, 1901 East
Madison Street, Baltimore, Md.
PEARSE, PROF. A. S., Duke University, Durham, N. C.
PEEBLES, PROF. FLORENCE, California Christian College, Los Angeles,
Calif.
PHILLIPS, DR. E. F., Cornell University, Ithaca, N. Y.
PHILLIPS, DR. RUTH L., Western College, Oxford, O.
PIKE, PROF. FRANK H., 437 West 59th St., New York City, N. Y.
PINNEY, DR. MARY E., Milwaukee-Downer College, Milwaukee, Wis.
PLOUGH, PROF. HAROLD H., Amherst College, Amherst, Mass.
POLLISTER, DR. A. W., Columbia University, New York City, N. Y.
42 MARINE BIOLOGICAL LABORATORY
POND, DR. SAMUEL E., University of Pennsylvania. School of Medicine,
Philadelphia, Pa.
PRATT, DR. FREDERICK H., Boston University, School of Medicine, Bos-
ton, Mass.
RAFFEL, DR. DANIEL, Johns Hopkins University, Baltimore. Md.
RAND, DR. HERBERT W., Harvard University, Cambridge, Mass.
RANKIN, PROF. W. M.. Princeton University, Princeton, N. J.
REDFIELD, DR. ALFRED C., Harvard University Medical School, Boston,
Mass.
REESE, PROF. ALBERT M., West Virginia University, Morgantown,
W. Va.
REINKE, DR. E. E., Vanderbilt University, Nashville, Tenn.
REZNIKOFF, DR. PAUL, Cornell University Medical College, New York
City, N. Y.
RHODES, PROF. ROBERT C., Emory University, Atlanta, Ga.
RICE, PROF. EDWARD L., Ohio Wesleyan University, Delaware, O.
RICHARDS, PROF. A., Unm-rsity of Oklahoma, Norman, Oklahoma.
RIGGS, MR. LAWRASON, JR., 25 Broad St., New York City, N. Y.
ROBERTSON, PROF. W. R. B., 1803 Anderson Street, Manhattan, Kan.
ROGERS, PROF. CHARLES G., Oberlin College, Oberlin, O.
ROMER, DR. ALFRED S., University of Chicago, Chicago, 111.
ROOT, DR. W. S.. Syracuse Medical School, Syracuse, N. Y.
SAMPSON, DR. MYRA M., Smith College, Northampton, Mass.
SANDS, Miss ADELAIDE G., 562 King St., Port Chester, N. Y.
SCHRADER, DR. FRANZ, Department of Zoology, Columbia University,
New York City, N. Y.
SCHRAMM, PROF. J. R., University of Pennsylvania, Philadelphia, Pa.
SCOTT, DR. ERNEST L., Columbia University, New York City, N. Y.
SCOTT, PROF. G. G., College of the City of New York, New York City,
N. Y.
SCOTT, PROF. JOHN W., University of Wyoming, Laramie, Wyoming.
SCOTT, PROF. WILLIAM B., 7 Cleveland Lane, Princeton, N. J.
SHULL, PROF. A. FRANKLIN, University of Michigan, Ann Arbor, Mich.
S HUM WAY, DR. WALDO, University of Illinois, Urbana, 111.
SIVICKIS, DR. P. B., Pasto deze 130, Kaunas, Lithuania.
SNOW, DR. LAETITIA M., Wellesley College, Wellesley, Mass.
SNYDF.R, PROF. CHARLES D., Johns Hopkins University Medical School,
Baltimore, Md.
SOLLMAN, DR. TORALD, Western Reserve University, Cleveland, O.
SONNEBORN, DR. T. M., Johns Hopkins University, Baltimore, Md.
SPEIDEL, DR. CARL C., University of Virginia, University, Va.
SPENCER, PROF. H. J., 24 West 10th St., New York City, N. Y.
REPORT OF THE DIRECTOR 43
STARK, DR. MARY B., New York Homeopathic Medical College and
Flower Hospital, New York City, N. Y.
STOCKARD, PROF. C. R., Cornell University Medical College, New York
City, N. Y.
STOREY, DR. ALMA G., Mount Holyoke College, South Hadley, Mass.
STRONG, PROF. O. S., College of Physicians and Surgeons, 630 West
168th Street, New York City, N. Y.
STUNKARD, DR. HORACE W., New York University, University Heights,
N. Y.
STURTEVANT, DR. ALFRED H., California Institute of Technology, Pas-
adena, Calif.
SUM WALT, DR. MARGARET, Women's Medical College, Philadelphia, Pa.
SWETT, DR. FRANCIS H., Duke University Medical School, Durham,
N. C.
TASHIRO, DR. SHIRO, Medical College, University of Cincinnati, Cin-
cinnati, O.
TAYLOR, Miss KATHERINE A., Cascade, Washington Co., Md.
TAYLOR, WILLIAM R., University of Michigan, Ann Arbor, Mich.
TENNENT, PROF. D. H., Bryn Mawr College, Bryn Mawr, Pa.
THATCHER, MR. LLOYD E., Canton, N. Y.
TINKHAM, Miss FLORENCE L., 71 Ingersoll Grove, Springfield, Mass.
TRACY, PROF. HENRY C., University of Kansas, Lawrence, Kans.
TREADWELL, PROF. A. L., Vassar College, Poughkeepsie, N. Y.
TURNER, PROF. C. L., Northwestern University, Evanstown, 111.
UHLEMEYER, Miss BERTHA, Washington University, St. Louis, Mo.
UHLENHUTH, DR. EDUARD, University of Maryland, School of Med-
icine, Baltimore, Md.
UNGER, DR. W. BYERS, Dartmouth College, Hanover, N. H.
VAN DER HEYDE, DR. H. C., Galeria, Corse, France.
VISSCHER, DR. J. PAUL, Western Reserve University, Cleveland, O.
WAITE, PROF. F. C., Western Reserve University Medical School,
Cleveland, O.
WALLACE, DR. LOUISE B., Spelman College, Atlanta, Ga.
WARD, PROF. HENRY B., University of Illinois, Urbana, 111.
WARREN, DR. HERBERT S., Department of Biology, Temple University,
Philadelphia, Pa.
WARREN, PROF. HOWARD C., Princeton University, Princeton, N. J.
WENRICH, DR. D. H., University of Pennsylvania, Philadelphia, Pa.
WHEDON, DR. A. D., North Dakota Agricultural College, Fargo, N. D.
WHEELER, PROF. W. M., Museum of Comparative Zoology, Cambridge,
Mass.
WHERRY, DR. W. B., Cincinnati Hospital, Cincinnati, O.
44 MARINE BIOLOGICAL LABORATORY
\YHITAKER, DR. DOUGLAS M., Columbia University, New York City,
N. V.
WHITE, DR. E. GRACE, Wilson College, Chambersburg, Pa.
WHITING, DR. PHINEAS W., University of Pittsburgh, Pittsburgh, Pa.
WHITNEY, DR. DAVID D., University of Nebraska, Lincoln, Nebr.
WIEMAN, PROF. H. L., University of Cincinnati, Cincinnati, O.
WILLIER, DR. B. H., University of Chicago, Chicago, 111.
WILSON, PROF. H. V., University of North Carolina, Chapel Hill, N. C.
WILSON, DR. J. W., Brown University, Providence, R. I.
WOGLOM, PROF. WILLIAM H., Columbia University, New York City,
N. Y.
WOODRUFF, PROF. L. L., Yale University, New Haven, Conn.
WOODWARD. DR. ALYALYX E.. Zoology Department, University of
Michigan, Ann Arbor, Mich.
YOUNG, DR. B. P., Cornell University, Ithaca, N. Y.
YOUNG, DR. D. B., University of Maine, Orono, Me.
ZELENY, DR. CHARLES, University of Illinois, Urbana, 111.
THE RELATION BETWEEN CLEAVAGE AND TOTAL
ACTIVATION IN ARTIFICIALLY ACTIVATED
EGGS OF URECIIIS
ALBERT TYLER
(From the William G. Kerckhoff Laboratories of the Biological Sciences, California
Institute of Technology, Pasadena, California and the William G. Kerckhoff
Marine Laboratory, Corona del Mar, California)
It is generally assumed in most work on artificial parthenogenesis
that cleavage and development result when the initial response of the
egg to the artificial agent most closely resembles its response to the
sperm. The percentage of eggs that respond in this fashion varies,
of course, with the length of exposure to the artificial agent, presumably
reaching a maximum for the exposure producing the highest percentage
of activation.1
It would follow then that the cleavage-activation relation should
be such that as the percentage of activation increases, the percentage
of cleavage increases; that is, that the percentage of cleavage is
directly proportional to the percentage of activation. Although this
relation is practically always tacitly assumed in parthenogenesis
experiments, detailed data on this point are lacking. If, however,
exposures giving higher percentages of activation do not produce
increasing percentages of eggs whose response is most nearly like that
induced by the sperm, or if such eggs were not the ones which cleave
and develop, an entirely different cleavage-activation relation might
be expected. The determination of this relation is important, then,
in an analysis of the factors which determine whether or not an
artificially activated egg will cleave.
The variation of the percentage of activation with the length of
exposure to the artificial agent is in itself a highly interesting fact,
since it is not manifested in insemination of a normal batch of eggs
with normal sperm. This variation may be attributed to variability
in the amount of treatment necessary to activate a given egg, or, less
likely perhaps, to a variation in the time at which the change produced
by the activating agent reaches a given egg. Whatever its source,
the way in which the percentage of activation varies with the length
of exposure is useful in helping to elucidate the mechanism by which
the artificial agent activates the egg.
1 Any egg in which initial developmental changes have taken place will be termed
''activated" in this paper, regardless of maturation or cleavage.
45
46
ALBERT TYLER
In the parthenogenesis experiments on Urechis eggs a unique
relation between cleavage and activation was found, such that as the
percentage of activation increases, the percentage of cleavage decreases.
The variation of percentage activation with length of exposure was
found to give a particular type of distribution curve in certain of the
experiments. These results together with their interpretation are
presented in detail in this report.
MATERIAL AND METHOD
The eggs used in these experiments were those of the echiuroid,
Urechis catipo, described by Fisher and MacGinitie (1928). The
changes undergone by the egg upon normal fertilization, and upon
artificial activation, and the method used in activating the eggs were
described in detail in a previous publication (Tyler, 1931). Briefly,
it was found that dilutions of sea water ranging from 80 per cent to
distilled water were effective in activating the Urechis eggs.2 In order
to treat the eggs, a batch was transferred with as little sea water as
possible to a Stender dish containing a large volume of the hypotonic
solution. Samples were then removed after various intervals of time
to Syracuse dishes containing normal sea water. All the usual pre-
cautions in regard to contamination by sperm or foreign matter,
hypertonicity, etc., were taken.
TARLE I
Unfertilized Eggs Treated with Distilled Water, Temperature 21.8° C.
Length of Exposure
Activation
Cleavage of Activated Eggs
min.
per cent
t>fr cent
0.05
. 57.0
32.4
0.08
95.0
10.0
0.17
99.6
0.6
0.25
100.0
0.0
0.33
100.0
0.05
0.42
100.0
0.1
0.50
100.0
0.0
0.67
100.0
0.0
0.83
100.0
0.0
1.00
100.0
1.5
1.50
100.0
0.0
2.00
100.0
0.0
3.00
100.0
0.0
4.00
100.0
0.0
5.00
99.0
0.0
2 Eighty per cent sea water, for example, is made up of eight parts sea water and
two parts distilled water. The sea water used was always taken at the same height
of tide.
CLEAVAGE-ACTIVATION RELATION
47
The percentages of cleavage and of activation were based on
counts of at least three hundred eggs; frequently, especially for very
low or very high percentages of activation, a much larger number
were counted.
It was shown that two types of activated eggs appear as a result
of the treatment. One type is characterized by initial changes which
are indistinguishable from those induced by the sperm. In this type
the breakdown of the germinal vesicle, the rounding out of the inden-
tation, the elevation of the membrane, and the extrusion of polar
bodies occur in very much the same manner as when the egg is fertilized
by a sperm. The time relations for these various changes, allowing
for the time of exposure, compare very closely with the time schedule
of the same events in the fertilized egg. However, in spite of the
remarkable similarity in behaviour of this type of artificially activated
egg to that of the fertilized egg, none of the eggs divide.3
The other type of artificially activated egg departs widely in its
behaviour from that of the normal fertilized egg. The only visible
TABLE II
Unfertilized Eggs Treated with 20 Per Cent Sea Water, Temperature 22.0° C.
Length of Exposure
Activation
Cleavage of Activated Eggs
min.
per cent
per cent
0.08
25.0
69.2
0.17
58.1
58.7
0.25
92.0
31.4
0.33
98.8
17.3
0.50
100.0
3.1
0.67
100.0
1.0
0.83
100.0
4.2
1.00
100.0
0.5
1.17
100.0
0.1
1.33
99.5
2.2
1.50
100.0
0.1
1.67
100.0
0.0
1.83
100.0
0.0
2.00
100.0
0.0
2.50
100.0
0.0
3.00
100.0
0.0
3.50
100.0
0.0
4.00
100.0
0.0
5.00
100.0
0.0
7.00
100.0
0.1
10.00
100.0
0.3
15.00
100.0
0.6
20.00
100.0
1.5
40.00
100.0
0.0
3 In only three cases were eggs with two polar bodies seen to divide. The three
eggs proceeded only as far as the two-cell stage.
48
ALBERT TYLER
change that occurs in this type of egg within the first three-quarters
of an hour after treatment is the dissolution of the germinal vesicle.
The egg remains indented, no membrane elevation occurs, and no
polar bodies are extruded. After that time the eggs begin to round
up, and lift off membranes, but no polar bodies appear. Practically
all of the eggs of this type divide, the time of first division varying
from one hour and twenty minutes to about three hours. The eggs
which cleave and develop are thus the ones which show a poor initial
response to the treatment. In what follows, then, the percentage of
cleavage is practically identical with the percentage of "poorly
activated eggs," and the data on percentage of cleavage and of activa-
tion will also show the relation between the percentage of "imper-
fectly' and of "perfectly" activated eggs for various strengths of
hypotonic solutions.
THE VARIATION OF PERCENTAGE OF ACTIVATION AND OF CLEAV.M.I.
WITH LENGTH OF EXPOSURE FOR VARIOUS
DILUTIONS OF SEA WATER
Treatment icith Distilled Water
The action of distilled water is extremely rapid in causing activation
of the eggs. After 3 seconds' treatment, 57 per cent of the eggs
TABLE III
Unfertilized Eggs Trailed -cilh 30 Per Cent Sea Water, Temperature 22.1° C.
Length of Exposure
Activation
Cleavage of Activated Eggs
min.
per rent
per cent
0.17
3.8
37.5
0.33
59.8
13.8
0.50
99.7
1.4
0.67
99.6
0.9
0.83
100.0
(i.l
1.00
100.0
0.6
1.33
100.0
0.4
1.67
' 98.1
3.1
2.00
99.2
0.7
2.50
99.0
0.5
3.00
99.3
0.8
3.50
98.5
1.6
4.00
83.6
2.3
4.50
86.1
3.2
5.00
81.8
11.1
6.00
86.7
1.0
8.00
100.0
0.0
10.00
100.0
0.1
15.00
100.0
0.0
20.00
100.0
0.0
40.00
100.0
0.0
CLEAVAGE-ACTIVATION RELATION
49
become activated, and after 15 seconds all of the eggs are activated.
The results of one series of exposures are given in Table I. Another
series run in the same manner gave quite similar results. The per-
centage of the activated eggs that cleave (column three in the table)
is seen to drop very rapidly as the percentage of activation increases.
Thus, when 100 per cent activation is obtained, there is practically
no cleavage.
Cytolysis sets in after 2 minutes' exposure and reaches 90 per cent
at 5 minutes' treatment. The activated eggs in that range are
somewhat abnormal in appearance, having a relatively wide membrane
and forming blisters over the surface so that polar bodies are often
indistinguishable.
Treatment with Twenty Per Cent Sea Water
The action of 20 per cent sea water is less rapid than that of
distilled water. The results of one series are given in Table II.
TABLE IV
Unfertilized Eggs Treated with 40 Per Cent Sea Water, Temperature 22.0° C.
Length of Exposure
Activation
Cleavage of Activated Eggs
min.
per cent
per cent
0.17
0.0
0.0
0.33
52.1
31.4
0.50
94.0
13.5
0.67
97.8
2.3
0.83
99.7
0.9
1.00
99.9
0.1
1.33
100.0
0.1
1.67
100.0
0.2
2.00
99.6
0.0
2.50
99.0
2.4
3.00
98.8
1.1
3.50
98.3
1.6
4.00
97.5
3.2
5.00
85.3
6.4
7.00
88.0
0.9
10.00
99.9
0.4
15.00
100.0
0.0
20.00
100.0
0.0
40.00
100.0
0.0
Fifteen seconds longer treatment is required to give 100 per cent
activation than for the distilled water. The rise in percentage of
activation with time of exposure is again seen to be accompanied by
a drop in cleavage. No exceptions are seen in the first part of the
table and the ones occurring in the latter part are of small magnitude.
Cytolysis sets in after 4 minutes' exposure and reaches 70 per cent
4
50
ALBERT TYLER
after 15 minutes and 90 per cent after 20 minutes. Increasing num-
bers of abnormal eggs of the type described above are found in that
range. Two other series of experiments were run, and closely similar
results obtained.
TABLE Y
Unfertilized Eggs Treated with 45 Per Cent Sea Water, Temperature 21.0° C.
Length of Exposure
Activation
Cleavage of
Activated Eggs
Volume in
M3 X 10 -»
min.
0.17
per cent
0.2
Per cent
0.0
0.33
11.5
50.0
0.50
39.0
29.2
0.67
81.3
10.6
0.83
92.9
4.8
1.00
98.6
1.4
8.12
1.33
100.0
0.2
1.67
100.0
0.0
2.00
99.5
0.0
8.90
2.50
98.3
0.3
3.00
96.4
0.8
9.51
3.50
87.3
2.2
4.00
65.7
4.0
10.10
5.00
34.4
20.5
10.45
6.00
21.2
26.2
10.92
8.00
7.9
22.2
11.49
10.00
6.9
50.0
11.92
15.00
0.8
43.0
12.95
20.00
1.3
51.6
13.51
40.00
3.3
8.3
13.96
Treatment icith Thirty Per Cent Sea Water
With 30 per cent sea water the percentage of activation rises less
rapidly than with 20 per cent. The results again show that as the
percentage of activation increases, the percentage of cleavage de-
creases. Table III gives the results of one series. The percentage
of activation shows a slight drop after about one and one-half minutes'
exposure which becomes quite marked at 4 to 6 minutes' exposure.
But as the activation drops, the cleavage is seen to rise, so that at 5
minutes' exposure, where the activation has dropped to 82 per cent,
the cleavage has risen to 11 per cent. The activation then rises
again to 100 per cent and the cleavage drops to zero.
Cytolysis sets in after 6 minutes' treatment and reaches 50 per cent
after 40 minutes. The abnormal eggs referred to above again appear
in this range of exposures.
Four other series of experiments were run, at temperatures ranging
from sixteen to twenty-three degrees, and very similar results obtained.
CLEAVAGE-ACTIVATION RELATK )N
51
The inverse relation between cleavage and activation was evident in
each series. If every case in which an increase (or decrease) in
activation accompanied by an increase (or decrease) in cleavage to
the extent of at least one per cent is considered an exception, then out
of a total of eighty-one dishes there are seven exceptions.
FIG. 1. Variation of percentage activation (open circles), percentage cleavage
(solid circles) and mean volume of eggs (continuous curve) with length of exposure
to 45 per cent sea water. Data of Table V.
Treatment with Forty Per Cent Sea Water
The percentage of activation for eggs treated with 40 per cent
sea water rises less rapidly than for eggs treated with any of the
preceding dilutions. Table IV gives the results of one series of
exposures. The activation is seen to rise rapidly to 100 per cent,
drop more slowly to 85 per cent and return again to 100 per cent.
The percentage of cleavage of the activated eggs decreases as the
activation increases, and increases as the activation drops. The
inverse relation between cleavage and activation is thus again clearly
shown, only one exception occurring in the table, namely at the three
minute exposure, where a drop in activation is followed by a drop in
cleavage greater than one per cent.
At the 8 minutes' exposure there is 4 per cent of cytolysis, which
increases to about 30 per cent for the 40 minutes' treatment. The
abnormal eggs again occur in this range.
52
ALBERT TYLER
Three other series were run with 40 per cent sea water, totaling
forty-seven dishes. Out of these a total of four exceptions of magni-
tude greater than one per cent were obtained.
TAHLK VI
Unfertilized Eggs Treated with 50 Per Cent Sea Water, Temperature 21.5° C.
Length of Exposure
Activation
Cleavage of
Activated Eggs
Volume in
MJ X 10-s
min.
0.17
per cent
0.1
per cent
100.0
0.33
2.0
80.1
0.50
43.2
40.6
0.67
91.0
21.6
0.83
96.7
10.6
1.00
99.2
5.3
7.83
1.33
99.8
0.7
1.67
100.0
0.4
2.00
100.0
0.4
8.72
2.50
100.0
1.2
3.00
90.8
26.9
9.29
3.50
93.4
21.9
4.00
86.0
39.2
9.72
4.50
68.3
63.2
5.00
33.9
76.2
10.10
6.00
34.9
80.0
10.46
8.00
18.8
94.7
11.08
10.00
7.1
83.3
11.42
15.00
0.5
40.0
12.20
20.00
0.4
79.0
12.72
40.00
1.5
61.3
12.85
Treatment with Forty-five Per Cent Sea Water
The results obtained with 45 per cent sea water differ in two
respects from those obtained with the preceding dilutions of sea
water. These are, first, that the percentage of activation returns
practically to zero after its initial rise to 100 per cent, and second,
that very little cytolysis sets in.
In Table V the results of one series of experiments are presented.
The rate of increase in activation is slower than with the preceding
dilutions. After 8 minutes' exposure a few of the eggs become cyto-
lysed and the amount of cytolysis reaches 5 per cent after 40 minutes.
The inverse relation between cleavage and activation is quite
evident in the table and is illustrated graphically in Fig. 1. The
"exceptions" generally occur in the dishes showing low percentages
of activation. Similar results were obtained in three other series of
experiments run with 45 per cent sea water. Out of a total of seventy-
CLEAVAGE-ACTIVATION RELATION
53
three dishes examined, sixteen exceptions were found, all of them in
dishes showing less than 8 per cent activation.
The increase in activation occurs much more rapidly than the
decrease. This may readily be seen in the graph (Fig. 1), where the
percentage of activation plotted against time gives a skew curve.
The probable interpretation of this result will be presented later.
But in connection with the activation-time curve, it is of interest to
90
BO
-tl-
II-J ~50
20
50%
FIG. 2. Variation of percentage activation (open circles), percentage cleavage
(solid circles) and mean volume of eggs (continuous curve) with length of exposure
to 50 per cent sea water. Data of Table VI.
present here the curve showing the increase in volume with length
of exposure to the 45 per cent sea water. The data from which the
curve was drawn are given in Table V. Each point represents the
average of the volumes of three eggs. The measurements of the
diameters were made with a Filar ocular micrometer. With this
micrometer measurements accurate to 0.1 per cent may be obtained.
However, the variations in volume, for the data presented here and
below, ranged as high as 5 per cent. This is probably due to the
rapid change in volume that is taking place as the measurements are
made and to the variability of the eggs. The volume measurements
are being repeated on a larger scale and by means of a cinematograph
in order to obtain accurate data for an analysis of the swelling process
itself. But even the relatively rough data presented here will be
shown to be useful in an analysis of the activation-time curves obtained
in these experiments.
54
ALBERT TYLER
The swelling curve of Fig. 1 shows that the eggs continue to
increase in volume even after the percentage of activation begins to
drop. The curve itself is of the exponential type, the slope continually
decreasing. In other words, the increase in volume occurs less rapidly
as the time of exposure increases.
TAHI.K VI I
Unfertilized Eggs Treated with 55 Per Cent Sea Water, Temperature 212° C.
Length of Exposure
Activation
Cleavage of
Activated Eggs
Volume in
M1 X 10-*
min.
0.17
per cent
0.4
per rent
100.0
0.33
0.9
100.0
0.50
10.0
53.3
0.67
38.4
37.4
0.83
70.3
29.4
1.00
89.5
17.6
7.74
1.33
96.2
' 9.6
1.67
98.4
2.0
2.00
99.9
1.0
8.73
2.50
97.8
1.6
3.00
93.3
5.6
9.32
3.50
88.8
7.1
4.00
85.6
12.4
9.57
5.00
66.4
21.3 .
9.82
6.00
64.1
21.0
10.11
8.00
47.3
16.6
10.49
10.00
13.9
15.4
10.89
12.00
1.7
44.2
15.00
1.1
70.6
11.41
20.00
0.0
0.0
11.63
40.00
2.2
90.9
11.84
Treatment with Fifty Per Cent .Sea Water
The results obtained with 50 per cent sea water are quite similar
to those obtained with 45 per cent sea water, except that the increase
in activation occurs more slowly and practically no cytolysis occurs
in any of the dishes. The increase in volume of the eggs in 50 per cent
sea water is also somewhat slower than for those exposed to 45 per cent
sea water, and the equilibrium volume attained is, of course, smaller.
Table VI contains the results of one series of experiments and a
set of volume measurements (means of three eggs) obtained at dilterent
times. The data are presented graphically in Fig. 2. The results
again bear out the inverse relation between cleavage and activation.
It is interesting to note that the irregular rise in the activation curve
CLEAVAGE-ACTIVATION RELAT I ON
55
at the three and one-half minutes' exposure is accompanied by a
drop in cleavage. The exceptions occur only in the last four points
of the graph, where the percentage of activation is low. A total of
eighty-two dishes in five series of experiments gave sixteen exceptions
all of the same type illustrated here.
The activation shows a drop to practically zero after its initial rise
100 «
Minutes
FIG. 3. Variation of percentage activation (open circles), percentage cleavage
(solid circles) and mean volume of eggs (continuous curve) with length of exposure
to 55 per cent sea water. Data of Table VII.
to 100 per cent, as in the preceding case. The rise in activation is
again seen to occur more rapidly than the subsequent decrease, giving
the skew activation-time curve shown in Fig. 2. The volume curve
for the eggs in the 50 per cent sea water is of the same type as obtained
with the preceding dilution.
Treatment with Fifty-five Per Cent Sea Water
The results of one series of experiments with 55 per cent sea water
and the volume data for the same dilution are presented in Table VII
and Fig. 3. The data again show a decrease in cleavage as the
activation increases and an increase in cleavage as the activation
decreases. Four series of experiments totaling seventy-four dishes
gave twelve exceptions — chiefly at low percentages of activation.
56
ALBERT TYLER
The activation-time curve is of the same shape as that obtained
in the preceding case, hut it is shifted slightly to the right, so that
the time required for the maximum percentage of activation and for
the return to zero per cent activation is longer than with 50 per cent
sea water.
The swelling curve (Fig. 3) shows that the volume continues to
increase after the percentage of activation has reached a maximum.
It is also of the exponential type in which the rate of increase in
volume decreases with time.
Treatment with Sixty Per Cent Sea Water
The results obtained with 60 per cent sea water again differ from
the preceding only in the time relations of activation and cleavage
and the volume curve. The data is given in Table VIII, and graphi-
cally represented in Fig. 4.
TAHLK VIII
Unfertilized Eggs Treated with 60 Per Cent Sea Water, Temperature 22.3° C.
Length of Exposure
Activation
Cleavage of
Activated Eggs
Volume in
M3 X 10-*
m i n .
ffr i '-ill
per cent
0.17
0.1
0.0
0.33
1.0
0.0
0.50
1.0
50.0
0.67
13.4
33.3
0.83
30.3
25.8
1.00
61.3
25.9
7.88
1.33
73.1
13.7
1.67
88.7
7.7
2.00
99.4
0.4
8.68
2.50
100.0
0.1
3.00
100.0
0.2
9.16
3.50
97.6
2.9
.
4.00
88.2
9.3
9.45
5.00
85.7
5.5
9.69
6.00
75.0
30.6
10.01
7.00
42.4
46.5
8.00
16.0
61.5
10.16
10.00
14.7
81.2
10.49
12.00
9.5
37.1
15.00
5.1
72.3
10.94
20.00
0.0
0.0
11.10
40.00
0.0
0.0
11.11
CLEAVAGE-ACTIVATION RELATION
57
The cleavage-activation relation shows up quite clearly. Out of
eighty-nine dishes in five series of experiments, fourteen relatively
unimportant exceptions were obtained.
ioor
60%
Minutes
FIG. 4. Variation of percentage activation (open circles), percentage cleavage
(solid circles) and mean volume of eggs (continuous curve) with length of exposure
to 60 per cent sea water. Data of Table VIII.
The activation-time curve shows a slight shift to the right when
compared with the preceding ones, but its asymmetry is still quite
evident.
The swelling curve is of the same type as in the preceding cases
but approaches a lower equilibrium volume.
Treatment with Sixty-Jive Per Cent Sea Water
Sixty-five per cent sea water gives results which differ from the
preceding in the same direction as the results obtained with the
60 per cent sea water differ from those obtained with 55 per cent.
One series of experiments and a set of mean volumes are shown in
Table IX and Fig. 5.
The inverse relation between cleavage and activation is evident
in spite of certain relatively large irregularities. From three series of
experiments seven relatively large and eight minor exceptions were
obtained out of a total of fifty-nine dishes.
58
ALBERT TYLER
TABLK IX
Unfertilized Eggs Treated with 65 Per Cent Sea Water, Temperature 22.0°
C.
Length of Exposure
Activation
Cleavage of
Activated Eggs
Volume in
M3 X 10-*
min.
0.17
per cent
0.0
per cent
0.0
0.33
0.2
0.0
0.50
0.0
0.0
0.67
0.7
40.0
0.83
6.4
36.3
1.00
13.6
11.8
7.61
1.33
77.8
20.8
1.67
87.6
13.4
2.00
92.0
10.2
8.32
2.50
96.5
6.1
3.00
98.6
2.1
8.73
3.50
92.5
6.7
4.00
71.9
22.5
9.25
5.00
49.0
56.7
9.56
6.00
21.8
46.9
9.49
7.00
20.3
51.0
8.00
13.8
42.4
9.88
10.00
13.7
34.9
10.15
12.00
6.0
40.9
15.00
3.9
15.3
10.50
20.00
0.2
100.0
10.55
40.00
0.0
0.0
10.55
The activation-time curve (Fig. 5) is asymmetrical as in the pre-
ceding cases, but it shows a slight shift to the right.
•
-
'
j|
•
ii-s tso
10-
9-0- 30
65%
Minutes
I i<. 5. Variation of percentage activation (open circles), percentage cleavage
-<>li<| circle^ ami mean volume of eggs (continuous curve) with length of exposure
in <o |.IT cent sea water. Data of Table IX.
CLEAVAGE-ACTIVATION RELATION
59
The volume increase is slower than for the eggs in more dilute sea
water, and approaches a lower asymptotic value.
Treatment with Seventy Per Cent Sea Water
The results of one series run with 70 per cent sea water are tabulated
in Table X. Figure 6 shows that trend of the data graphically.
TABLE X
Unfertilized Eggs Treated with 70 Per Cent Sea Water, Temperature 212° C.
Length of Exposure
Activation
Cleavage of Activated Eggs
min.
per cent
per cent
0.17
0.4
57.1
0.33
0.8
50.0
0.50
2.5
55.2
0.67
2.4
29.8
0.83
4.1
30.2
1.00
4.6
29.5
1.33
6.4
20.3
1.67
20.4
19.5
2.00
46.1
10.2
2.50
96.2
9.1
3.00
100.0
2.3
3.50
97.5
10.9
4.00
69.2
19.7
4.50
51.6
35.7
5.00
31.6
55.5
6.00
16.7
65.2
8.00
7.4
82.9
10.00
13.3
76.3
15.00
12.7
85.8
20.00
4.6
83.3
40.00
0.3
66.7
No serious divergence from the cleavage-activation relation is
evident. Two series of experiments totaling thirty-seven dishes gave
seven minor variations.
The activation-time curve (Fig. 6) is again decidedly asymmetrical.
It is displaced to the right, so that the return to zero per cent activation
requires a longer exposure than in the preceding cases.
The volume data are not presented for this or for the succeeding
dilutions of sea water. The volume increase proceeds more slowly,
of course, and reaches a smaller equilibrium volume with increasing
concentrations of sea water.
60
ALBERT TYLER
100
•
•
•
40
a. 30
.-
6 . , 7 8
Mi nutes
12
FIG. 6. Variation of percentage activation (open circles) and percentage cleav-
age (solid circles) with time of exposure to 70 per cent sea water. Data of Table X.
Treatment icith Seventy-fire Per Cent Sea Water
Table XI and Fig. 7 contain the results of one series of experiments
with 75 per cent sea water.
TABLE XI
Unfertilized Eggs Treated with 75 Per Cent Sen \\',iter, Temperature 20.8° C.
Length of Exposure
Activation
. .f Ai tiv.it'M Eggs
min.
per cent
/vr cent
0.17
0.0
0.0
0.33
0.0
0.0
0.50
0.0
0.0
0.67
0.0
0.0
0.83
0.0
0.0
1.00
0.0
0.0
1.33
3.3
71.4
1.67
8.2
50.0
2.00
33.9
45.9
2.50
83.2
I'). 5
3.00
86.2
32.3
3.50
90.0
28.4
4.00
79.3
5.00
47.8
59.2
6.00
30.2
65.0
8.00
7.4
61.3
10.00
1.2
66.7
15.00
0.1
40.0
20.00
0.0
0.0
CLEAVAGE-ACTIVATION RELATION
61
As before, the percentage of cleavage varies inversely with the
percentage of activation, although the difference between the maximum
of activation and the corresponding minimum of cleavage is not as
great as in the cases listed above. Two series of experiments totaling
thirty-one dishes gave four exceptions.
lOOr
Minutes
FIG. 7. Variation of percentage activation (open circles) and percentage cleav-
age (solid circles) with time of exposure to 75 per cent sea water. Data of Table XI.
For the activation-time curve (Fig. 7), the time to reach a maximum
is longer than in the preceding case, but the drop to zero per cent
occurs sooner. However, the maximum value reached is only 90 per
cent activation as compared with 100 per cent in the previous cases.
The curve itself is still asymmetrical.
Treatment with Eighty Per Cent Sea Water
Eighty per cent sea water generally fails to give more than one
to two per cent activation. In one series of experiments, however,
an exceptionally high percentage of activation was obtained. The
results are given in Table XII and Fig. 8.
It is readily seen from the data that practically every increase
(or decrease) in activation is accompanied by a decrease (or increase)
in cleavage, bearing out the inverse relation between cleavage and
activation.
62
ALBERT TYLER
The activation-time curve reaches its maximum as quickly as for
the 75 per cent sea water, but that is undoubtedly due to the higher
temperature at which this series was run. The activation curve does
not return to zero, but maintains a relatively high percentage of
activation and a correspondingly high percentage of cleavage.
TABLE XII
Unfertilized Eggs Treated with 80 Per Cent Sea Water, Temperature 22.5° F.
Length of Exposure
Activation
Cleavage of Activated Eggs
min.
per cent
per cent
0.17
0.0
0.0
0.33
0.0
0.0
0.50
0.0
0.0
0.67
0.0
0.0
0.83
0.2
100.0
1.00
1.4
42.8
1.33
6.6
31.2
1.67
14.5
30.8
2.00
58.2
26.8
2.50
94.1
21.5
3.00
95.7
17.6
3.50
98.8
3.5
4.00
98.7
8.3
5.00
81.6
57.8
6.00
92.5
37.1
8.00
80.7
48.6
10.00
65.5
59.1
15.00
64.1
38.2
20.00
62.5
48.8
40.00
47.6
64.2
THE VARIATION OF PERCENTAGE OF ACTIVATION WITH VOLUME
FOR VARIOUS DILUTIONS OF SKA WATER
When (he percentage of activation is plotted against the mean
volume attained by the eggs at different lengths of exposure, a curve
is obtained which is much more symmetrical than the activation-time
curve. Figure 9 shows five curves of that type, for 45, 50, 55, 60
and 65 per cent sea water. The percentages of activation were plotted
in each case against the volumes attained at corresponding times of
exposures, the volumes being taken from the smooth curves.
The activation-volume curves of Fig. 9 approach in shape the
normal distribution curve. The individual curves have the same
abscissa but the ordinates are raised successively for each dilution of
i water. It can readily be seen that even with the same coordinates
CLEAVAGE-ACTIVATION RELATION
63
the curves would not coincide; but their divergence is no greater than
would be expected when one considers the statistical nature of the
activation values and the errors involved in the volume measurements.
Moreover, there are probably injury factors operative in the lower
concentrations of sea water that are not present in the higher concen-
trations, as indicated by the cytolysis obtained in 45 per cent sea
water.
FERTILIZATION OF "OVER-EXPOSED" EGGS
The activation-time curves for concentrations of sea water above
40 per cent are seen to rise to a maximum of about 100 per cent
activation and then drop off to zero.
100
90
80
It
60
2 &50
1 i
o -2
<t o
I
«£ 30
20
80%
QlOOCXX
Minutes
age
FIG. 8. Variation of percentage activation (open circles) and percentage cleav-
(solid circles) with time of exposure to 70 per cent sea water. Data of Table XII.
The eggs which do not respond before the "optimum exposure"
is reached may be termed "under-exposed" unactivated eggs, and
those which do not respond upon longer exposures may be termed
"over-exposed" unactivated eggs.
The failure of the "over-exposed" unactivated eggs to respond to
the treatment might presumably be due to an injury effect, or other
change produced in the eggs. The "over-exposed" unactivated eggs
as well as the under-exposed unactivated eggs were therefore insemi-
nated with fresh sperm in order to determine whether they would
become fertilized and produce normal embryos. The results obtained
64
ALBERT TYLER
with 45 to 65 per cent sea water are given in Table XIII. The third
column in the table gives the total percentages of activation obtained
with the len.uths of exposure listed in column two. The fourth column
gives the percentage of the unactivated eggs that become fertilized
upon addition of sperm, and the fifth column, the percentage of the
fertilized eggs that produce normal larvae.
45%
FIG. 9. Variation of percentage of activation with mean volume of eggs at-
tained at corresponding times of exposure to 65, 60, 55, 50, and 45 per cent sea water.
Ordinates raised successively for each dilution of sea water. Data from Tables Y to
IX; volumes taken from the smooth curves of Figs. 1 to 5.
The unactivated eggs were transferred to a separate dish and
inseminated at about 2 to S hours after treatment. Control eggs
(listed in the table as 0.0 minutes' exposure) were inseminated at the
same time.
The "under-exposed" unactivated eggs are not given for the 55
and the 60 per cent sea water. In the other three cases the "under-
exposed" unactivated ei^s ^ho\v practically 100 per cent fertilization
and a high percentage of normal embryos. The "over-exposed" eggs
show a high percentage of fertilization in every case, comparing quite
favorably with that given by the control eggs. The percentage of
normal embryos obtained varies considerably, but is quite as good as
that obtained from the controls, except for the 45 per cent sea water.
I lo\\<-\ c] -, in the latter case a relatively large percentage of the eggs
were pol\>perniic.
CLEAVAGE-ACTIVATION RELATION
65
TABLE XIII
Insemination of" Under-Exposed" and "Over-Exposed" Unactivated Eggs
Concentration of
Sea Water
Length of
Exposure
Ai t iv;iti<m
Fertilization
Normal Embryos
per cent
min.
f>?r i'fnt
per cent
per cent
0.0
•
95
40
0.17
0.0
99
50
1.50
100.0
— •
—
45
10.00
5.7
70
20
15.00
0.0
60
2
20.00
0.3
60
5
40.00
0.5
60
5
0.0
—
98
25
50
0.17
0.1
100
60
2.00
100.0
—
—
15.00
0.5
80
70
0.0
—
65
70
55
2.00
100.0
— .
. — .
20.00
0.0
50
65
0.0
100
100
2.75
100.0
—
—
60
7.00
42.4
100
100
10.00
14.7
100
95
20.00
0.0
100
100
0.0
100
65
0.17
0.0
100
100
65
3.00
99.0
—
—
7.00
28.1
100
40
10.00
18.3
90
75
20.00
0.0
99
50
The results show that the "over-exposed" unactivated eggs are
still capable of becoming fertilized, even though a shorter exposure
would have resulted in every egg becoming activated upon return to
normal sea water.
DISCUSSION
1 . Variation of Rate of Increase in Activation with Dilution of Sea-Water
It is evident from the results presented above that the factors
causing activation are brought into action more quickly, the lower
the concentration of the sea water used for the treatment. In the
dilute sea water the egg swells due to intake of wrater. The volume
increase also occurs more quickly, the lower the concentration of the
sea water in which the eggs are allowed to swell. This parallel be-
haviour suggests that volume increase in the dilute sea water may be
5
66 ALBERT TYLER
used as a basis for an interpretation of the results presented above;
but this is not meant to imply that water-intake alone is responsible
for the activation of the egg.4
2. Activation-Time Curces
For concentrations of sea water ranging from 45 to 75 per cent the
percentage of activation was seen to rise rapidly to a maximum and
then fall off more slowly. In terms of volume change this means
that when the egg is in a definite volume range it will become activated
upon return to normal sea water, but before or after passing through
that volume range the egg does not become activated upon return to
normal sea water. This volume range is evidently well below the
equilibrium volume, since the eggs continue to swell after the time
of exposure giving the maximum activation. The reason why a
range of volumes rather than one definite volume is specified will be
indicated below. There is considerable variability in the time at
which different eggs pass through the same volume range when a
given batch is exposed to a given dilution of sea water. Thus some
of the eggs will have reached the volume range from which return to
normal sea water results in activation before the others have entered
that range. Correspondingly, some of the eggs will have passed
through that volume range while the others are still in it. Let us
term the volume range resulting in activation the "optimum volume
range." The percentage of eggs passing through a given volume
range at a given time will depend on the kind of variability shown by
the eggs. If the variability of this material is expressed by the
normal distribution curve, then we would expect the variability in
the percentage of eggs passing through the " optimum volume range "
to be expressed by that type of curve only if the increase in volume
were a linear function of the time of exposure. But the volume
increase is a logarithmic function of time, the rate of swelling continu-
ally decreasing with time of exposure. The eggs therefore enter the
"optimum volume range" more rapidly than they leave it. Thus the
variation in the percentage of eggs passing through the "optimum
volume range" with time of exposure should be expressed by a skew
distribution curve with its mode displaced to the left. In other
words, the variation of percentage activation with time of exposure
should give a skew curve, since the percentage of eggs passing through
the "optimum volume range" is by definition identical with the
percentage of activation. This is in fact the type <>t curve that is
4 The change in hydrogen ion concentration, for example, might be an important
factor. It ranged from pH 8.2 for the sea water to pi I 7.1 for the distilled water used.
CLEAVAGE-ACTIVATION RELATION 67
obtained when percentage of activation is plotted against length of
exposure (Figs. 1 to 8).
The reason for assuming a range of volumes rather than one definite
optimum volume results from the following consideration. The
maximum of the activation-time curve is at 100 per cent activation.
This means that all of the eggs must be in such a condition after a
certain time of exposure that removal to normal sea water at that
time results in every egg becoming activated. But the volume
measurements show that the eggs vary in the time of exposure at
which a given volume is reached. Therefore, if we adhere to the
volume interpretation we must assume that a range of volumes, at
least as great as the variation in volume of the individual eggs, is
effective in causing activation upon return to ordinary sea water.
The time of exposure at which all of the eggs are in that "optimum
volume range" then results in 100 per cent activation.
On this basis the more rapid swelling obtained with progressively
lower concentrations of sea water should cause a shifting of the
activation-time curve to the left proportional to the increase in rate
of swelling and likewise a shortening of the time range of activation.
The results presented above show that this is in general true. But
with extreme dilutions of sea water (40 per cent to distilled water)
the drop to zero per cent activation does not occur. This is probably
due to a secondary effect as indicated by the fact that there is a
tendency for the activation to drop (see Tables II, III, IV), but as
cytolysis sets in a second rise in activation (of an abnormal type)
takes place.
One should also expect, according to the volume interpretation,
that the concentration of sea water in which the equilibrium volume
of the eggs is within the "optimum volume range" should give an
activation-time curve that does not drop. This is presumably
approached by the 80 per cent sea water (Table XII and Fig. 8).
3. Activation- Volume Curves
If the variation of the percentage of activation with time of
exposure is correlated with the variation in volume of the eggs attained
at corresponding times of exposure, then the percentage of activation
plotted against mean volume should give a normal distribution curve,
which should be identical for the various dilutions of sea water.
The results show that this is roughly true. The curves obtained with
various dilutions of sea water (Fig. 9) are quite symmetrical when
compared with the activation-time curves. The probable reasons
for the failure of the various curves to be exactly identical have been
given above.
68 ALBERT TYLER
The expectation of a normal distribution curve for percentage
activation plotted against mean volume is based on the assumption
that the variation in volume of the eggs, at each time of exposure
considered, is expressed by the normal probability curve. This is the
type of variation that is generally assumed for biological material in
the absence of further information. To obtain such information in
this case it would be necessary to measure the volumes of a large
number of eggs at various times of exposure. This has not been done
on a large enough scale and accurately enough to determine whether
the chance law holds for the volumes at every exposure used, but the
measurements obtained on untreated eggs indicate that their variation
in volume is of that type.
4. " Over-Exposed" Unactiva ted Eggs
It has been shown that the over-exposed unactivated eggs obtained
with solutions ranging from 45 per cent to 65 per cent sea water can
still be fertilized and may produce normal embryos. This may be
taken to mean that the eggs have not been irreversibly affected by
treatment with these dilutions of sea water. Consonant with this
fact is the observation previously reported, that no visible changes
aside from the swelling are seen to occur in the treated eggs while in
the dilute sea water. It is also in accord with the result that the time
for the initial stages (e.g. polar body extrusion) of the artificially
activated eggs is comparable with that of the fertilized eggs only if
allowance is made for the time of treatment.
It is evident then that no developmental changes occur in the egg
while in the hypotonic solution, but that activation is initiated by the
return to normal sea water after a definite time of exposure (or after
a certain amount of water has been taken in). The question may
therefore be raised as to why a longer exposure fails to evoke a response
in the egg upon return to normal sea water when a shorter one does.
If the egg were found to be injured by the longer exposure this question
might be more readily answered. But the data presented here show
that this is not so. The question bears directly on the mechanics of
activation. With the data available we can only answer by restating
the result in the following terms — that a definite change (enabling the
egg to become activated upon return to normal sea water) is produced
in the egg by the intake of an amount of water within a certain range,
but that the change is reversed when more water is taken in. In
other words, by the difference in behaviour upon return to normal
sea water, an egg in the optimum exposure range must be intrinsically
different from an egg in the earlier or later ranges, and by the similarity
CLEAVAGE-ACTIVATION RELATION- 69
in behaviour upon return to normal sea water, an egg in the earlier
range of exposures must be intrinsically the same (neglecting the
manifest difference in volume) as an egg in the later range; hence the
change produced must be reversed.
The return to original condition of eggs that have been allowed
to swell in dilute sea water has also been noted in eggs of Nereis
(Just, 1930) and eggs of Arbacia (McCutcheon and Lucke, 1926).
But in neither of these cases is it stated whether activation is obtained
at shorter exposures.
The ability of eggs that have been "over-exposed" to butyric acid
to become fertilized has been noted by Moore (1916) for Arbacia,
Just (1919) for Echinarachnius, and Lillie (1921) for Strongylocentrotus.
But in these cases the cleavage and development were stated to be
abnormal.
5. Cleavage- Activation Relation
The inverse relation between the percentage of cleavage and the
total percentage of activation may now be interpreted in a similar
way provided we introduce a "sub-optimum volume range" on both
sides of the "optimum volume range." The justification of this arises
from a consideration of the results reported in a previous publication
(Tyler, 1931). It was shown that the activated eggs that extrude
both polar bodies practically never divide, even though the response
of that type of egg to the treatment is outwardly indistinguishable
from the response of the egg to the sperm. Only the eggs that produce
no polar bodies were the ones to cleave, but such eggs were shown to
respond in a relatively very slow and abnormal fashion to the treatment
in respect to the breakdown of the germinal vesicle, rounding out of
indentation, and membrane elevation. Such is the type of result one
would expect from a "sub-optimum" treatment. In terms of volume
change this "sub-optimum" exposure would be obtained in a "sub-
optimum volume range." Practically no eggs of that type are
obtained at the time of exposure giving 100 per cent activation, but
they occur in increasing numbers to either side of that exposure time.
Since, at the time of exposure giving 100 per cent activation, all of
the eggs are assumed to be in the "optimum volume range," the
"sub-optimum volume range" must occur on each side of the former.5
Thus, when a batch of eggs is treated with dilute sea water, the
eggs will pass through a "sub-optimum volume range" both before
and after entering the "optimum volume range." At relatively short
5 The "sub-optimum volume range" must evidently be shorter than the range of
variability of the volumes of the eggs, since 100 per cent cleavage (with 100 per cent
activation) is never obtained for any given exposure.
70 ALBERT TYLER
times of exposure, then, one would expect most of the activated eggs
to be within the "sub-optimum volume range," and so give a high
percentage of cleavage (of the activated eggs). But with longer
exposures as the total activation increases one would expect more and
more of the eggs to enter the "optimum volume range" and so give
a low percentage of cleavage. The results would then be reversed
upon passing through the second "sub-optimum volume range" with
longer exposures.
This leads to a relation between percentage of cleavage and
percentage total activation that is identical with that described in
the text.
This interpretation can be tested in a much better fashion by
following the volume changes of individual eggs in various dilutions
of sea water and noting their behaviour when removed to normal
sea water after having been allowed to swell to various volumes.
Such experiments are now in progress.
The results reported here have an important bearing on what is
generally termed the "optimum treatment" in parthenogenesis
experiments. It has generally been assumed that the treatment
producing the highest percentage of activation (similar to that pro-
duced by the sperm), and of cleavage and development is the optimum
treatment. But in Urechis it has been shown that the treatment
that is optimum for activation is not so for cleavage and development.
Thus, if one wishes to produce the most parthenogenetic development,
the length of exposure used is different from that which would be
chosen if one wished to produce the highest percentage of eggs whose
initial response to the treatment was most similar to that induced by
the sperm. It is preferable, I think, to term the latter the optimum
treatment, for the reasons stated above. The failure of eggs receiving
the optimum treatment to divide is probably connected with insuffi-
cient chromatin (since all such eggs extrude two polar bodies and are
left with the haploid number of chromosomes). It should be possible
then to produce cleavage in such eggs by suppressing the polar divi-
sions. This is somewhat difficult to accomplish without initiating
other changes in the eggs, but the resu'ts obtained thus far indicate
that suppression of the polar divisions of the "optimally" stimulated
eggs results in cleavage.
The inverse relation between percentage of cleavage and percentage
of activation appears then to depend on the fact that only the "poorly
activated eggs" which extrude no polar bodies are the ones to divide.
Thus the extent to which this relation is general for eggs of various
forms will probably depend on whether or not the eggs that extrude
CLEAVAGE-ACTIVATION RELATION 71
both polar bodies divide. In eggs of the sea urchin type, where the
polar bodies are extruded in the ovary and where cleavage is apparently
possible with the haploid number of chromosomes, we might not
expect this relation to hold.
In eggs of Thalassema neptuni, which, from the descriptions are
very similar to the Ureclris eggs, artificial activation by means of
isotonic solutions has been reported by Hobson (1928). The varia-
tions of percentage of activation and of percentage of cleavage c are
presented for several short series of exposures, but Hobson thinks '
that the results show an increase in cleavage with increase in activation.
However, he notes (pp. 73 and 74) that the maximum of cleavage
often fails to coincide with the maximum of activation, when both
composition of medium and length of exposure are varied.
SUMMARY
1. The rate of increase in percentage activation of Urechis eggs
with hypotonic sea water is shown to decrease as the concentration
of sea water used is increased from distilled water to 80 per cent
sea water.
2. The rate of increase in volume also decreases with increased
concentration of sea water.
3. For dilutions of sea water ranging from 75 per cent to 45 per
cent, the activation passes through a maximum (usually 100 per cent)
and then returns to zero per cent with longer exposures. For lower
concentrations of sea water the return to zero per cent is not obtained,
but a high percentage of activation is maintained. With 80 per cent
sea water the return to zero per cent activation also does not occur.
4. The activation-time curves for 75 per cent to 45 per cent sea
water are of the form of skew distribution curves, rising rapidly to
100 per cent activation and falling more slowly to zero per cent.
5. The activation-volume curves are presented for 65 per cent to
45 per cent sea water and are of the form of a normal probability curve.
They are roughly identical for the various dilutions of sea water.
6. Practically every series of experiments shows an inverse relation
between the percentage of total activation and percentage of cleavage
(of the activated eggs) ; so that as the percentage of activation increases
with time of exposure, the percentage of cleavage decreases, and when
6 Hobson's total activation does not include cleavage. It is not stated in the
paper whether the percentage of cleavage is that of all the eggs or of the activated
eggs, though it seems to be the former. When the data of his tables is recalculated on
this basis, there are thirteen cases in which an increase (or decrease) in activation is
accompanied by an increase (or decrease) in cleavage and six cases in which the in-
verse relation holds.
72 ALBERT TYLER
the percentage of activation decreases with exposure the percentage
of cleavage increases.
7. The over-exposed unactivated eggs are still capable of fertiliza-
tion and of producing normal embryos in spite of the fact that a shorter
exposure would have resulted in their becoming activated upon return
to normal sea water.
8. The variation in rate of activation with concentration of sea
water, the type of activation-time curves, the activation-volume
curves, and the fertilization of over-exposed eggs are shown to be
interpretable on the basis of volume change occurring in the dilute
sea water, a definite volume range being optimum for activation.
The cleavage-activation relation is shown to be the outcome of the
previously reported result that only the "poorly activated" eggs
divide, and its interpretation, based also on the exposures producing
such eggs, involves the assumption of a "sub-optimum volume range"
on both sides of the- optimum.
BIBLIOGRAPHY
liMiMK, \V. K. AND (>. E. MAC i.iMMK, 1(>28. A new Echiuroid Worm. Ann. and
Mag. Xat. Hist., Ser. 10, 1: 199.
FISHKU, \V. K. AND G. E. MAC I.IMIIK, 1<)2,X. Tin- Natural History of an Echiuroid
Worm. Ann. anil Mag. Xul. Hist., Ser. 10, 1: 204.
HOBSOX, A. D., 1928. The Action of Isotonic Salt Solutions on the Unfertilized
Eggs of Thalassema neptuni. Writ. Jour. Exper. Biol., 6: 65.
|i ST, E. I\.. \')\(). The Fertili/ation Reaction in Echinarachnius parma. III. The
nature of the activation of the egg by butyric acid. Biol. Bull., 36: 39.
Ji ST, E. I-'.., 1930. ilydration and Dehydration in the Living Cell. III. The
fertilization capacity of Nereis eggs after exposure to hypotonic sea water
Protoplnsmu, 10: 24.
LILLIE, E. K., 1921. Studies of Eertili/at ion. IX. On the question of superposition
of fertili/ation on parthenogenesis in Strongylocentrotus purpuratus. Biol.
Hull., 40: 23.
McCu'K in <>N.. M., AND I.ICKK, B., 1926. The Kinetics ot ( >smoiic Swelling in Liv-
ing Cells. Jour, di'ii. Physiol., 9: 0(>7.
MOORE, C. K., 1916. On the Superposition of Kertiii/ation on Parthenogenesis.
Hi«l. Hull., 31: IS 7.
, ALBKKI. I'MI. The Production of Normal Embryos by Artificial Partheno-
genesis in the Hrliiumi.l. I'rechis caupo. Biol. Bull., 60: 187.
THE OCCURRENCE OF MELANOPHORES IN CERTAIN
EXPERIMENTAL WOUNDS OF THE GOLDFISH
(CARASSIUS AURATUSY
GEORGE MILTON* SMITH
ANATOMICAL LABORATORY, SCHOOL OF MEDICINE, YALE UNIVERSITY
While studying in the goldfish the repair of experimental wounds,
crushes, burns, and fractures, it became apparent that melanophores
developed in the wounds a few days after the trauma and later de-
generated and thus disappeared. Not alone did these melanophores
occur directly at the site of the injury, but not infrequently in the
corium of adjacent areas and even in remote cutaneous regions. In
none of these places were black pigmented cells seen by a previous low-
power microscopic examination of the living fishes used for the experi-
ment, nor were melanophores of the corium noticeable by high magnifi-
cation in sections of tissue removed from the region of the wound at
the time of trauma. The appearance of pigmented cells at the very
point of injury seemed to indicate a role of importance for melanophores
of this fish, from the viewpoint that these cells functioned in the pro-
cesses of repair and, not unlikely, in the mechanism of body defense.
As results of different experiments were found to be uniform, only a
few are here reported in detail as illustrative.
Experiment 1. Goldfish, 8 cm. long from snout to base of tail,
kept in still water tank, supplied by current of air. Temperature of
Wc.ter 78° F.
Oct. 28, 1930. Transverse incision was made with a cataract knife
through a single ray of caudal fin, near the upper edge of middle part
of this fin. Incision penetrated tissues over both surfaces of fractured
ray.
Oct. 30. Overlying the ray near the fracture are a few scattered
melanophores with irregular processes (Fig. 1). Tissues overlying the
fracture are cedematous and difficult to photograph for this reason.
There are a few small points of hemorrhage near the fractured frag-
ments.
Oct. 31. A large number of melanophores, in places interlacing,
surround the proximal and the distal fragment of the fractured ray as
if to encapsulate the fragments (Fig. 2).
1 Aided by grant from Blossom Fund.
73
74 GEORGE MILTON SMITH
Xov. 1. Active degeneration of melanophores has begun with
pigment granules lying free in tissue spaces (Fig. 3).
Xov. 5. Degeneration of all melanophores in the region of the
fracture, with many small pigment masses scattered throughout the
field.
Xov. 8. Entire region of fracture, somewhat whitish and trans-
lucent, shows no more evidence of pigment.
In the following experiment multiple injuries were produced.
Experiment 2. Two goldfishes, 7 cm. from snout to base of tail
were placed in a tank of still water fed with a current of air. The
temperature of the water was gradually raised from 70° F. by heating
over a period of three days to 84° F.
Sept. 23, 1930. In both fishes eight different regions were clamped
with an artery forceps each for 15 seconds. The points clamped were
as follows: right and left opercuhun, both pectoral, both ventral, the
anal and the caudal fins.
Sept. 25, 2 P.M. One fish shows early pigmentation by melano-
phores in caudal fin, the second fish has melanophores in the right ven-
tral fin. Pigmentation is slightly distal to crush.
Sept. 26, 3 P.M. Three days after trauma, both fishes show pig-
mentation by melanophores at all eight points crushed. The pigmen-
tation is a marked one due to the large number of melanophores present
in the crushed zones and neighboring tissue.
EXPLANATION OF PLATE 1
FIGS. 1-4. Experimental linear fracture by incision of a ray of caudal fin of
goldfish, the injury including all tissues directly overlying fracture. Letters A and B
indicate site of fracture. All photomicrographs taken from the same living fish
ana-sthetised with chloretone 1-201)0. Magnification X 90. Temperature of water
78°-80° F.
Kn;. 1. Two days after injury. A few melanophores have appeared in the
u-dematous tissue near the fracture, A.B.
FIG. 2. Three days after injury. Numerous melanophores appearing as single
cells or interlacing cells at the line of fracture, A.B.
IK.. 3. Four days after injury. Degeneration of melanophores at the site of
fracture A.B. has begun. Small black pigment masses from degenerated cells lie
scattered among living melanophores.
In.. 4. Five days after injury. Degeneration of melanophores at the site of
fracture A.B. is complete, scattered pigment debris remains in the field. Final dis-
appearance of all pigment on the eighth day after injury.
I n.. 5. Inter-radial tissue of caudal fin showing melanophores distributed near
capillaries marked A, B, C, D. X 60. Fresh tissue removed from goldfish near an
area crushed eight days previously. Fish outdoors exposed to sunlight,
FH,. 6. Irregular areas of pigmentation of melanophores developing on the
surface of the body of a goldfish injured by removal of all body scales five days
previously. Photograph made from living fish anaesthetised with chloretone.
Size, two-thirds normal.
MELANOPHORES IN THE ( .< )U >KISI 1
75
I'l \TE I
2.
3.
Lr
5.
6.
76 GEORGE MILTON SMITH
Sept. 28. There is evidence of degeneration of melanophores at all
crushed points. Temperature, 90° F.
Oct. 5. One fish is entirely clear of degenerated pigment granules
at crushed point. The second shows a few black granules in the wound
of the caudal fin.
Oct. 6. In both fishes all evidence of pigment formed of de-
ici.iU'd melanophores has disappeared at all eight points crushed.
Thus these two fishes injured by crushing at eight separate points have
shown, with temperature of water between 84° F. and 90° F., an intense
pigmentation by melanophores at crushed points, a subsequent de-
generation of melanophores, and a complete disappearance of all pig-
ment detritus all in the course of 13 days.
Experiment 3. In this experiment, involving injury to the right
operculum, 30 goldfishes, about seven cm. in length, were used. These
were divided into three groups of ten. Each group was placed in a
separate tank of running water in the laboratory. Fishes in Tank 1
were operated on by resecting one third of the right operculum by a
straight vertical cut with scissors. Fishes in Tank 2 received a simple
vertical crush for fifteen seconds of the middle of the right operculum.
Fishes in Tank 3 were tirst crushed tor 15 seconds by a clamp placed
vertically in the mid-point of the operculum and all opercular tissue
distal to the clamp was resected. Tank 1 — Fishes (simple excision of
one half of the right operculum) showed melanophores in the margin of
the wound three days after operation. At first only a few such cells,
but in the following two or three days there were many present.
Evidence of degeneration of melanophores was noted in places as early
as two days after their first appearance. Complete disappearance of
black degenerated pigment from the wounded area varied between 3
and 9 days. Fishes in Tanks 2 and 3 with more severe injuries of the
operculum showed a beginning accumulation of melanophores in the
injured operculum also three days after trauma. The entire disap-
pearance of pigment from the wound in fishes in Tank 2 (vertical crush
of operculum) varied between (> to 15 days after appearance of melano-
phores. In Tank 3 (fishes with crushed and partially resected right
operculum) the eruption of melanophores at the injury occurred also
three days after injury, bin the final disappearance of pigmented
debris varied between 9 and 16 days. One fish in Tank 1 and four
fishes in Tank 2 showed slight pigmentation by melanophores of the
opposite uninjured operculum, arising when the accumulation of
melanophores on the injured side was well developed. Melanophores
in the area of secondary pigmentation degenerated and disappeared
those in the experimentally injured right operculum.
MKLANOPHORES IN THE GOLDFISH
SUMMARY
The onset of cutaneous pigmentation by melanophores in three
different types of wound of the operculum carried on simultaneously in
three different tanks of running water at 76° F. was uniformly between
the third and fourth day after trauma. The final disappearance of
pigment of degenerated melanophores of the wound area varied between
6 and 19 days after injury. In some fishes the accumulation of melano-
phores noted at the wound was relatively slight; in others the black
pigmentation caused by large numbers of melanophores was intense and
remained over a longer period.
Fishes operated on during the cold winter months and kept in tanks
of cold running water (43° F.) did not show at wounded areas such a
rapid development of melanophores as described in the preceding ex-
periment. Further, pigmentation of wounds under winter temperature
extended over longer periods. Thus, in nine fishes with right oper-
culum crushed for 15 seconds with an artery clamp placed at the middle
of the operculum, followed by excision of opercular tissue distal to the
operculum, the following results were obtained: An eruption at the
injured operculum in all nine fishes occurred between 13 to 16 days
after injury; pigmentation had cleared up by degeneration of melano-
phores in only three fishes two months after injury, with temperature
of water at 53° F. It took approximately one more month (tempera-
ture 53°-56° F.) for four more fishes to clear; the remaining two fishes
cleared at the end of still another month or four months from the date
of injury, when the temperature of the water had gradually risen to
61° F. The longest period of pigmentation in a wound of this series
represented approximately 110 days from the date of the first appear-
ance of melanophores.
It became of interest to learn whether or not in fishes kept in very
cold water, an appearance of melanophores after trauma could be
temporarily inhibited, to appear for the first time when such fishes
were changed back slowly to more favorable warmer temperatures.
A number of experiments were done along these lines.
Experiment 4. A goldfish, seven cm. in length, was placed in a
tank of still cold water supplied by a current of air, the water varying
in temperature between 42° F. and 45° F. The tank was set up in a
refrigerator arranged with a double window, admitting ample daylight.
It was found advisable to accustom the experimental fishes gradually
to cold. By using several submerged electric lights at the beginning of
the experiment and turning these off as desired, the temperature of the
water could be lowered slowly without endangering the life of the fish.
78 GEORGE MILTON SMITH
Oct. 14, 1930. A small incision was made with a cataract knife in
the caudal tin of this goldfish dividing transversely a single ray near
the upper margin of the tin. Examination of melanophores at four
day intervals negative for an entire month. Temperature 42° to 46° F.
Nov. 14. Temperature in tank raised slowly so as to reach 66° F.
on Nov. 16th.
Nov. H). Numerous melanophores appeared for the first time at
fracture and along injured ray distal to this. No other black pigmenta-
tion noted.
Nov. 25. Slight pigmentation by melanophores of tip of tail and
also along the margin of dorsal fin. Large accumulation of melano-
phores at fracture.
Nov. 27. Active degeneration of melanophores at fracture and
other pigmented regions.
Nov. 29. Fish under dissecting microscope shows no pigment
masses either at site of experimentally fractured ray or at the second-
ary points of black pigmentation of tail or dorsal tin. All melano-
phores have disappeared by a process of degeneration.
Fishes kept in a dark chamber, excluding all light, developed mel-
anophores in wounds as promptly as did controls kept in daylight.
Experiment 5. Two goldfishes, seven cm. in length, with crushed
right operculum and caudal fin, kept in a dark chamber in a tank of
still water at 64° F., supplied by air current, were taken out of this
chamber to be examined for the first time after injury on the fifth day.
Many melanophores were present in crushed regions. At the same
time, two control fishes, injured on the same day in a similar way, kept
in a tank of equal size at the same temperature but exposed to labora-
tory daylight, exhibited, also for the first time, a large number of melan-
ophores at the two crushed points. Twenty-three days after injury,
one fish contained in the dark chamber and both controls were clear of
pigment; the second fish in the dark chamber showed no melanophores
in the injured operculum, although a few small masses of degenerated
pigment masses still remained in the caudal fin.
The production of a second injury in a healed wound frequently,
but not always, caused another eruption of melanophores. Refractur-
ing a single ray at the same point, especially where the previous healing
had left a wide whitish translucent area, did not produce a second crop
of melanophores. The very simple injury of making a longitudinal
slit in the caudal fin did not call forth melanophores either at the time
of the first injury or with repeated incisions at the same point.
The irregular topographic distribution of melanophores following
MELANOPHORES OK THE GOLDFISH 79
trauma was seen particularly well in experiments where the scales on
both sides of the body were totally removed.
Experiment 6. Nov. 1930. Three goldfishes, A, B, C, measuring
8, 7, 5 cm. in length respectively, kept in a heated tank of still water
76° F., supplied by air current, were operated on under chloretone
anaesthesia (1-2000). All scales of the body were removed with
forceps in all three fishes.
Dec. Four days after operation, melanophores appeared in ir-
regular groups at various points on both sides of the body. The two
larger fishes, A and B, showed in the course of the next few days a large
number of melanophores in irregular scattered patches. The patches
of pigmentation by melanophores in fish B are shown in Fig. 6. The
smallest fish, C, showed only a few melanophores in small, widely-
scattered areas. By the end of the twelfth day degeneration of melan-
ophores evoked by removal of scales had occurred in all three fishes
with a disappearance of broken-down pigment material. At this time
(12 days after removal of scales) each fish showed definitely a set of new
young scales. Fish B successfully withstood a second complete re-
moval of scales, under chloretone anaesthesia, but this time only a very
few rapidly degenerating melanophores developed on the denuded
surface of the body, as if the supply of pigment-forming cells for these
particular surface areas were partially exhausted. When, however, on
the fourth day after the second operation for removal of scales the
caudal fin of this fish was crushed by clamp for 15 seconds, numerous
melanophores developed three days later in the crushed tail but in no
other place.
DISCUSSION AND SUMMARY
Various important problems relating to melanophores and melano-
genesis appear in connection with the works of Van Rynberk (1906),
von Frisch (1911), Weidenreich (1912), Asvadourova (1913), Spaeth
(1913), R. Fuchs (1914), Wyman (1924), Wells (1925), Abolin (1925),
Ewing (1926), Jost (1926), Bloch (1927), Cordier (1928), Becker (1930).
For the present purpose it may be of interest to recall that a number
of years ago Weidenreich (1912) showed that in vertebrates the distri-
bution of black pigment cells could be regarded as forming four
distinct envelopes for the body. These envelopes he designated as
"cutaneous, perineural, pericoelomatic and perivascular" respectively.
He pointed out that whereas in some vertebrates several or all of these
pigmentary envelopes were well developed, in other vertebrates one or
more of these pigmentary envelopes might be found poorly developed,
showing only a trace or rudiment of pigmented tissue. For example,
80 GEORGE MILTON" SMITH
in man, where there exists a. well developed cutaneous envelope of pig-
mented tissue, the perineural pigmented tissue is poorly developed,
presenting itself as scattered black pigment cells of the piamater and
elsewhere in the brain. In fishes all pigmentary envelopes are regarded
as fairly well developed.
In interpreting the meaning of melanophores following injury as
seen in the above experiments on goldfish, it should be kept in mind
that such melanophores may represent a perivascular or perineural
type of cell developing the properties of forming pigment, rather than
cells belonging strictly to a system of cutaneous melanophores. It is
particularly the cutaneous or corial melanophores of fishes which have
received the most study to date.
Melanophores, according to Bloch (1927) show a number of mor-
phologic peculiarities in that they form processes or dendrites and have
a tendency to arrange themselves in an interlacing net\vork. They
exhibit in cold-blooded animals certain functional reactions which are
shown by the spreading or the contraction of the intracellular masses of
pigment granules. These reactions are changes which have their
origin in nervous, actinic or hormonal stimuli; and they may also be
produced by mechanical, chemical and electric means.
Ever since the description of melanophores in fishes by Siebold
(1861), many investigators have contributed to the morphology of this
subject. The works of Ballowitz (1912-16) on the different types of
chromatophores (i.e., the melanophores, xantho or erythrophores,
^iianophores and their various combinations forming what he desig-
nated as chromatic organs) have largely laid the basis for our present
knowledge of pigment cells in fishes. This author also demonstrated
histologically the innervation of melanophores in fishes.
The experimental observations of Pouchet (1876) showed a rela-
tionship between cutaneous melanophores in fishes and the sympathetic
nervous system. It remained, however, for von Frisch (1911, 1912),
in a series of important experiments, to demonstrate in fishes a con-
traction center for cutaneous melanophores in the front part of the
medulla, and a secondary center in the spinal cord. Further, he ex-
plained the pathways by which impulses pass from brain through pig-
ment motor nerve fibers to the sympathetic system and from here by
means of the peripheral nerves not only to the melanophores but also
to other chromatophores of the skin.
In general, the function of melanophores has been variously in-
terpreted. In addition to the view that cutaneous pigmentation and
pigment changes represent color adaptation to environment, the pur-
pose of cutaneous pigment has been thought to lie in its protection ot
MELANOPHORES IX THE GOLDFISH 81
deeper tissues against injurious solar rays. The migration of retinal
pigment granules as it applies to vertebrates and arthropods is thought
by Parker (1906) to be a mechanism calculated to protect the receptive
organs of the retina from o~\ e -stimulation by light and to improve the
' retinal images. Cutaneous pigment cells have been regarded as trans-
forming light into heat energy. According to this view, as Weidenreich
(1912) explains, the minute individual intracellular pigment granules of
melanophores become heat bodies or Heizkorper, which distribute heat
to neighboring protoplasm. Weidenreich (1912) has further suggested,
because melanophores are innervated and react to optic, thermic and
chromatic stimuli, that they may be regarded perhaps as sensory cells
for color and warmth perception.
Cordier (1^28) believes that the formation in cells of melanin is a
process of excretion as yet not well understood. The theory implies
that certain toxic waste products of metabolism gain access to special
cells and there become insoluble and pigmented, their toxic products
being neutralized. Elimination of pigment follows slowly as if it were
a process of retarded excretion. Certain clinical cases of Addison's
disease and melanosarcoma have shown melanin greatly increased in
cutaneous areas and present in the blood and in the urine. This has
been taken to mean a profound chemical disturbance of the body as a
whole and gives support to the view that a general metabolic process
may ordinarily affect the production of melanin in various regions of
the body.
Whatever may be the relationship to the nervous system of melano-
phores resulting from trauma as seen in the present experiments on
goldfish, it seems plausible from their structural arrangement in healing
wounds, that such melanophores are pigmented cells which function in
repair of damaged tissue. Melanophores of this kind appeared rela-
tively early in the course of wound-healing when favorable warm tem-
peratures were employed. They disappeared by a process of degenera-
tion at the site of the wound when healing proceeded and usually when
the covering of the wounded surface was nearing its completion.
Whereas melanophores showed in wrounds of goldfishes within 3 or 4
days after injury when fishes were kept in water of relatively warm
temperature (70°-90° F.), with fishes kept in cold water (40°-42° F.)
the appearance of melanophores in wounds was retarded or even in-
hibited, to appear for the first time when these fishes were returned
to a warm environment. A temperature of 40° F. was found sufficient
to inhibit the appearance of melanophores for a month.
Fishes kept in a dark chamber completely excluding light showed
melanophores in various experimental wounds as early as did controls
6
82 GEORGE MILTON SMITH
kept under usual laboratory conditions exposed to light. Fishes kept
in tanks out-of-doors and in this way exposed directly to the sunlight
developed melanophores in wounds a few days later (Fig. 5). The
reaction here seemed intense. In some of these fishes melanophores
developed not alone at the crushed points, but also in areas adjacent to
the wound and in all other fins.
When studied in a simple form of injury such as dividing trans-
versely a single ray of the caudal fin, melanophores appeared first as
periadventitial cells in close relation to the outer \valls of the small
capillary blood vessels which covered the surface of the ray near the
fracture. With an increase in numbers, the melanophores spread
toward the region of the fracture and formed a network (Fig. 2) in the
corium by the interlacing of the numerous irregular processes. De-
generation in individual melanophores was observed as early as 24
hours after their first appearance near a fractured ray. Fixed paraffin
sections of tissue with degenerating melanophores showed a moderate
number of phagocytic cells containing pigment. For the most part,
however, the impression was gained that the pigment detritus rested
free in the tissue spaces preparatory to removal by lymphatics, or
became dissolved in situ.
The actual production of melanin in cells is now generally regarded
as the result of enzyme action. The important studies of Bloch (1927),
advancing the views on the intracellular production of melanin by
enzyme, are too well known to need repetition here. It is conceivable
that in the experimental wounds of goldfish chemical changes occur
locally permitting melanin to be formed in periadventitial cells ir-
regularly distributed in the corium of the injured area.
Experimental wounds of goldfishes quite naturally are constantly
open to infection by bacteria or parasites. Numerous bacteria and
especially cocci were seen in paraffin sections of tissue from crushed
operculum at various stages after injury before complete healing had
occurred. When, as occasionally noted, a growth of fungus appeared
in connection with experimental wounds, pigmentation by melano-
phores appeared particularly intense, affecting not alone the wound
but also adjacent areas. There was at times pigmentation of the fins
other than the ones experimentally injured and, in rare instances, a
patchy pigmentation of body scales under these circumstances.
Treating such wounds for several days in succession with two per cent
mercurochrome destroyed the parasites, and pigmentation of the
wound with secondary pigmented areas then disappeared. The pres-
ence of bacteria in wounds and the large number of melanophores
present in injured areas affected with parasites, suggest a possible role
l"r melanophores in the mechanism ot body defense.
MELANOPHORES IX THE GOLDFISH
Goldfishes subjected to a total removal of scales showed in the
course of several days a distribution of melanophores varying in extent
and intensity in different fishes. This eruption was asymmetrical, ir-
regular and patchy, as if periadventitial cells capable of forming black
pigment as a result of trauma or during subsequent wound regeneration,
actually occupied a very irregular distribution on both sides of the
body. As new scales formed in these experimentally produced scale-
less fishes, melanophores disappeared by degeneration. A second total
removal of scales in one of the fishes was followed by a very scanty
eruption of melanophores, as if the possibility of local melanophore
production in this instance were, temporarily, at least, exhausted.
Usually, but not always, a re-injury at the same point brought out a
second eruption of melanophores differing but little from that which
followed the primary injury.
The eruption of melanophores in experimental wounds of the gold-
fishes, varying in intensity in different fishes, appears to indicate that
such melanophores, probably periadventitial in origin, form in response
to injury and function in the repair of injured tissues.
LITERATURE CITED
ABOLIN, L., 1925. Beeinflussung des Fischfarbenwechsels durch Chemikalien.
Arch. f. mikr. Anat. und Entwick., 104: 667.
ASVADOUROVA, N., 1913. Recherches sur la formation de quelques cellules pigmen-
taires et des pigments. Arch, d'anat. micros., 15: 153.
BALLOWITZ, E., 1893. Die Nervenendigungen der Pigmentzellen, ein Beitrag zur
Kenntnis des Zusammenhanges der Endverzweigungen der Nerven mit dem
Protoplasma der Zellen. Zeitschr.f. wissenschaft. Zool., 56: 673.
BECKER, S. W., 1930. Cutaneous Melanoma: a Histologic Study especially directed
toward the Study of Melanoblasts. Arch. Dermal, and Syph., 21: 818.
BLOCH, B., 1927. Das Pigment. Handbuch d. Haut. u. Geschlechtskrankheiten
Berlin, Vol. 1, Part 1, pp, 434-541.
CORDIER, R., 1928. Les pigments melaniques et la melanogenese. Bull. Soc. Roy.
d. Sc. med. e nat. de Bruxelles, Nos. 2-7, pp. 43-57.
EWING, J., 1922. Neoplastic Diseases. Philadelphia, pp. 871-890.
VON FRISCH, K., 1911. Beitrage zur Physiologic der Pigmentzellen in der Fischhaut.
Arch.f. ges. Physiol., 138: 319.
VON FRISCH, K., 1912. Uber farbige Anpassung bei Fischen. Zool. Jahrbuch, 32:
171.
FUCHS, R. F., 1914. Der Farbenvvechsel und die chromatische Hautfunktion der
Tiere. Handbuch d. vergleich. Phys., 3: 1189.
Josx, F., 1926. Die Farbzellen und Farbzellvereinigungen in der Haut des Nordsee-
fisches Callionymus lyra L. Zeitschr.f. mikr. anat. Forschung, 7: 461.
PARKER, G. H., 1906. The Influence of Light and Heat on the Movement of the
Melanophore Pigment, especially in Lizards. Jour. Exper. Zool., 3: 401.
POUCHET, G., 1876. Des Changements de coloration sous 1'influence des nerfs.
Jour, de I' Anat. et de Physiol., 12: 1-90, continued 113-165.
VON SIEBOLD, C., 1863. Die Siisswasserfische von Mitteleuropa. Leipzig, p. 14.
SPAETH, R. A., 1913. The Physiology of the Chromatophores of Fishes. Jour.
Ex per. Zool., 15: 527.
84 GEORGE MILTON SMITH
VAN RYXBERK, G., 1906. t'ber den durch Chromatophoren bedingten Farbenwechsel
der Tiere (sog. chromatsche Hautfunktion). Ergebn. der Physiol., 5: 347.
WEIDENREICH, F., 1912. Die Lokalisation des Pigmentes und ihre Bedeutung in
Ontogenie und Phylogenie der \Yirbeltiere. Zeitschr. f. Morph. u. Anthrop.,
Sonderheft 2, pp. 59-140.
WEILS, H. G., 1925. Chemical Pathology. Philadelphia, pp. 526-532.
WY.MAX, L. C., 1924. Blood and Nerve as controlling Agents in the Movements of
Melanophores. Jour. Exper. Zool., 39: 73.
STUDIES ONr THE INTESTINAL FLORA OF TERMITES
WITH REFERENCE TO THEIR ABILITY TO
DIGEST CELLULOSE
ALBERT DICKMAX
(From the Department of Bacteriology, University of Pennsylvania)
INTRODUCTION
Interest for a considerable time has been centered on the ability of
certain organisms to derive nourishment from a wood diet, the principal
constituents of which are cellulose and lignin, both resistant to the
digestive action of enzymes normally present in the digestive tract of
most animals. Animals such as termites, larvae of wood-boring
beetles, and a bivalve, the shipworm Teredo navalis, so injurious to
ships and piles, have been conspicuous for their ability to digest
cellulose. In the case of Teredo navalis (Dore and Miller, 1923),
digestion has been explained by the production of cellulose-digesting
enzymes. The explanation in the case of termites is a more interesting
one.
Microscopic examination of the intestinal content of most species
of termites discloses countless numbers of Protozoa and bacteria.
Careful experiments have been carried out to explore the relationships
between the host and their intestinal organisms, and it has been shown
conclusively that the termites are absolutely dependent upon the
Protozoa present for the digestion of the cellulose in their food.
L. R. Cleveland, one of the foremost workers in this field, has shown
that termites containing an intestinal fauna of Protozoa were able,
under favorable conditions, to exist on a diet entirely made up of
Whatman filter paper of the purest grade, and he successfully reared
them upon this diet for over two years. If, however, he defaunated
the termites (by incubating them at 36° C., the Protozoa were killed,
but the termites were unharmed), they were unable to live upon the
pure cellulose and soon died. If he re-inoculated the termites with
Protozoa, however, after incubation, they were able to live indefinitely
upon the filter paper (Cleveland, 1924).
BIOLOGICAL CONSIDERATIONS
Comparatively little work has been done with respect to the
biological relationship between termites and the other intestinal organ-
isms usually present. Hollande (1922) discusses the morphology and
85
86 ALBERT DICKMAX
reproduction in a considerable number of spirocructes which he found
very abundant in the intestinal contents. Hoelling (1910), in his
paper on "The Xuclear Conditions of Fusiformis termitidis," describes
the morphology of fusiform bacilli studied from a number of smears of
tlu' intestinal content of termites. Imms (1924) states that " Portier
in\ c-n-, ited an apparent symbiosis in the case of the larva of Nonagria
which lives within the stems of Typha devouring the pith. ... In the
digestive tube of this larva are found great numbers of motile conidia
of a fungus (Isaria), which exist among the devoured vegetable frag-
ments. The conidia are always accompanied by a micrococcus which
secretes an enzyme capable of dissolving cellulose. Portier states that
the conidia develop at the expense of the dissolved cellulose and
eventually penetrate the walls of the gut, escaping into the blood.
Most of them are there attacked by phagocytes and transformed into
products which serve to nourish the tissue of the host."
In an attempt to determine whether the Protozoa were entirely
responsible for the digestion of cellulose in the digestive tract of
termites, Cleveland (1924) studied the bacterial flora of Reticulitermes
flavipes. He states that bacteria were sometimes numerous, and he
attempted all known methods, aerobic and anaerobic, for isolating
cellulose-digesting bacteria. One hundred attempts were made, but
all results were negative, even after the cultures were more than two
months old. Ten attempts to isolate cellulose-decomposing molds and
actinomyretes were made and were unsuccessful. In an attempt to
isolate the organisms, an inorganic, medium was made containing:
KoHPO, 1.00 gram
MgSO, 0.50 gram
KC1 0.50 gram
leSO, .01 gram
Na.\<>. 2.00 grams
H<) lOOOcc.
To this medium cellulose was added in two forms: a small piece of
Whatman's filter paper and 0.5 per cent cellulose suspension. To the
inorganic medium containing cellulose suspension sufficient agar was
added to make a solid medium. Incubation was apparently 36° C.
Cleveland (1(^2S) observed that all families of termites harbored
many spirocruetes which he thought might play a role in the digestion
o| cellulose and hemicellulose. He observed millions of these, often
attached to a single Proto/ofm, and easily mistaken for flagella.
At tempi- to -row the spirocruetes failed and animal inoculations
I n-ved negative. By feeding the termites cellulose thoroughly moist-
INTESTINAL FLORA OF TERMITES 87
ened with a 5 per cent aqueous solution of acid fuchsin, he found it
possible to remove in this manner all spirocha?tes without doing any
damage whatever to the Protozoa or to the termites. He concluded
that the spiroclurtes play little if any role in the digestion of wood and
cellulose.
In the wood-ingesting larvae of certain insects, characteristic blind
sacs and diverticula of the digestive tract have been demonstrated in
which, aided by the action of myriads of bacteria, food particles are
held and digested. From the larvae of rose-chafers (Potosia cuprea)
cellulose-digesting bacteria have been isolated in pure culture. These
slender peritrichiate, anaerobic rods (Bacillus cellulosae fermentans
\Yerner) are found free also in the ant hills inhabited by Potosia cuprea.
The optimum temperature for fermentation was found to be 33-37° C.,
the minimum 21° C. The larvae are so dependent upon these intestinal
organisms that the increase in weight of the larvae is determined by the
temperature. If the temperature of the ant hill at the end of October
goes below 21° C., the now useless taking-in of food material is sus-
pended (Buchner, 1928).
EXPERIMENTAL
a. The Food of Termites
It was the purpose of the present experiment to determine whether
cellulose-digesting organisms occurred in woody material upon which
termites feed. The woody material from termite colonies was first
examined for cellulose-digesting organisms. Some material obtained
from a termite colony in March, 1929, which had been kept in a dry
condition since that time (almost one year), was used to inoculate
nitrate-cellulose tubes. The material contained wood particles,
termite excreta and a small quantity of fine reddish clay.
The nitrate-cellulose medium was made according to the formula
of Bradley and Rettger (1927). It contained:
Di Potassium phosphate 1 gram
Magnesium sulfate 1 gram
Sodium chloride 1 gram
Calcium carbonate 2 grams
Potassium nitrate 2 grams
Distilled water 1000 cc.
The cellulose was provided in the form of strips of filter paper.
Tubes were kept at room temperature and incubated at 34.5° C. and
60.5° C. under aerobic and anaerobic conditions.
Ten tubes were incubated at room temperature under aerobic
88 ALBERT DICKMAX
conditions. In every one of these there was abundant growth and
discoloration of paper above the liquid level. Marked discoloration
appeared in all tubes in four days. Seven tubes labeled G-l to G-7
inclusive were inoculated from two of the above tubes on March 1 ,
1930. When examined on March 3, 1930 discolored areas were present
on the paper of all tubes. On the original tubes the discolorations
were yellow-green, yellow and light brown. On the "G" tubes the
predominant color was dark brown.
Three tubes inoculated and placed under anaerobic conditions on
February 19, when examined on March 8, showed no marked dis-
coloration and no cutting of paper at this time. (Anaerobic conditions
were produced in a Mason jar, using pyrogallic acid and sodium hy-
droxide.) On this date the three tubes were placed under aerobic
conditions. When examined April 7, tube no. 16 showed cutting of
paper at liquid level and maceration of paper below liquid level.
Mold growth apparently was inhibited.
A tube no. 12 incubated at 34.5° C. under aerobic conditions on
February 19, when examined on February 25 showed marked discolora-
tion similar in variety and extent to that produced at room tempera-
ture. The paper, however, was not cut at liquid level.
Three tubes 17, 18, and 19, inoculated and incubated at 34.5° C.
under anaerobic conditions, showed no noticeable discoloration or
cutting of paper when examined 18 days later. The tubes were re-
moved from the incubator and placed under aerobic conditions, at room
temperature. When examined 16 days later, the paper in tubes 18 and
19 was entirely cut at liquid level with no characteristic discoloration.
The paper below the liquid level was entirely macerated. (The paper
was probably cut before the day of examination.)
A tube incubated at 60.5° C". under aerobic conditions showed,
when examined about 10 days later, a few small, isolated areas of
growth on paper.
Three tubes were inoculated and incubated at 60.5° C. under
anaerobic conditions on February 20. When examined on March 7,
one of the tubes showed paper cut at liquid level and macerated at
lower portion below liquid level. In this tube the paper was so mace-
rated that upon slight shaking it fell apart into loose fibers below liquid
level.
From this last tube three tubes were inoculated and incubated at
60.5° C. under aerobic conditions on March 8, three silica-gel plates
inoculated from the above tube also, were incubated anaerobically at
60.5° C. On April 7, neither plates nor tubes showed discoloration or
growth. Re-inoculated nitrate-cellulose tubes at 60.5° C. under
anaerobic conditions showed no growth after one month.
INTESTINAL FLORA OF TERMITES 89
Woody material obtained with termites from Dr. Cory was used to
inoculate a tube at room temperature under aerobic conditions. Th
material was composed of wood and digested material, but no loos
soil. When examined 9 days later the paper at the liquid level showe^
a discoloration which was marked above liquid level. Green discolora-\
tion was prominent. One colony 0.5 mm. in diameter produced a wine-
colored discoloration. Transplants on silica-gel plates produced
abundant growth.
The silica-gel was made fundamentally according to the general
formula of Winogradsky (1929). Seventy-five grams of potassium
silicate were dissolved in 1000 cc. of distilled water. To this was
added an equal quantity of HC1 of a specific gravity of 1.10. Thirty
cc. of mixture were placed in the petri dishes and set aside under cover
for 24 hours. They were then immersed in running water for 24 hours.
They were next washed four successive times for 24 hours each in large
covered dishes of sterile distilled water. From the following inorganic
salt solution (made up for 100 plates) 2 cc. were added to each silica-gel
dish.
KH2PO4 1.0 gram
MgSO4 0.5 gram
NaCl 0.5 gram
FeSO4 0.01 gram
MnSO4 0.01 gram
KNO;! 3.6 grams
CaCO- 2.0 grams
Distilled water 200 cc.
The pH was adjusted to 7.2. The petri dishes were then placed in an
incubator at 60.5° C. until excess moisture evaporated. Sterilized
pieces of Whatman's filter paper were placed aseptically on silica-gel
plates. Sterile covers were then placed over the plates. Stroke in-
oculations were made.
Termites received from Dr. Cory were transferred to clean petri
dishes in which were placed filter paper and the tissue paper sent with
the original shipment. This was being eaten by the termites. On
March 8, pellets of termite excreta which had been dropped on the
tissue paper were used to inoculate two nitrate-cellulose tubes and one
silica-gel plate, and kept at room temperature. (The pellets were clean
and the color of tissue paper.) On March 10 no visible growth was
evident in the tubes. Yellow discoloration was noticeable on the paper
around the pellets in the silica-gel plates. This growth later covered
the entire paper and the paper showed almost entire digestion by May
10. Nitrate-cellulose tubes inoculated from this plate showed growth
90 ALBERT DICKMAX
and digestion of paper in 5 days. A gelatinous milky-white growth
with translucent areas developed on paper. Microscopic examination
showed maceration of fibers with attached bacteria. The growth was
characterized by the presence of mold growth which formed the
. i i.itinous milky mass.
b. The Intestinal Contents of Termites
Most of the experiments with the intestinal contents were carried
out with termites of the genus Reticulitermes collected at Mullica Hill,
Xew Jersey, and with Tennopsis, received from Dr. Kirby, at the
University of California.
Reticulitermes- Microscopic examination of intestinal contents
showed besides the myriads of Protozoa, spirocrurtes ranging in length
from 5 to 15 n; the smaller ones were in great abundance, and appar-
ently more numerous than the Protozoa. Motile rods and filamentous
rods were also present. Examinations were made with hanging drops
in physiological salt solution and from smears stained with alcoholic
fuchsin.
On March 31, seven inoculations were made with intestinal con-
tents of seven termites, washed 4 minutes in 1- 1000 HgClo, then rinsed
with sterile distilled water. The intestinal contents were squeezed out
with >tcrilc forceps onto silica-gel plates. These were kept anaerobi-
cally at room temperature. On April 1, sixteen termites were used as
above to inoculate silica-gel plates. Eight of these were kept under
anaerobic conditions and eight were kept at room temperature under
aerobic conditions. On April 2, twenty termites were used as above to
inoculate silica-gel plates kept at room temperature under aerobic
conditions. ( )n April 1 , six termites were used to inoculate six nitrate-
cellulose tubes.
When the above cultures were examined on May 12, none of the
anaerobic pl.iic- -Imwed growth. Of the aerobic silica-gel plates eight
inoculations showed .1 Alight mold growth, with no cutting ot paper
and no distinct discoloration. Tin- remaining inoculations on plates
produced no growth. One of the six nitrate-cellulose tubes showed
clearly digestion of paper, with translucent areas, and microscopic
examination of paper from this tube showed numerous bacteria,
1 X 0.75 /j. in si/e on the libers.
Four tubes inoculated on February 28 with Reticulitermes flavipes
received from Dr. L. K. Cleveland showed no growth on March 10,
and no growth was visible on April M).
1 i-nini/tsis. — The specimens were large enough to enable one to dis-
sect out tin- digestive tract with sterile forceps. All termites were
INTESTINAL FLORA OF TERMITES <H
first washed for 5 minutes in 1 1000 HgCl2 and then rinsed in sterile
distilled water. All inoculations were made on silica-gel plates,
which were kept under aerobic conditions at room temperature.
On April 21, eight termites were used for inoculations. On April
23, two winged termites were used to inoculate silica-gel plates. On
April 24, ten winged termites were used for inoculations. On April 24,
eight worker termites were used as above. On May 2, six workers were
used as above. On May 2, eight winged forms were used as above.
On April 23, ten termites were used to inoculate silica-gel plates which
were kept under anaerobic conditions. When examined on May 17,
none of the plates showed cellulose-digestion or indication of growth of
cellulose-digesting organisms. There was mold growth on intestinal
contents of eighteen termites. The growth of molds was evidently due
to the fact that treatment for 5 minutes with 1-1000 HgCl2 did not
kill them.
Beckwith and Rose (1929) obtained cellulose digestion in a number
of cases wThen working with intestinal contents of termites, but their
results cannot be taken as conclusive, since they attempted to sterilize
the termites externally by merely washing them in tincture of iodine
(U.S. P.) for 45 seconds. This short exposure and the fact that small
air bubbles captured between the hairs of the insect would prevent
contact with the germicide in this time, would indicate that the organ-
isms on the surface were not destroyed.
DISCUSSION OF RESULTS
Cleveland and others have shown that digestion of cellulose in
termites is entirely dependent upon intestinal organisms. Cleveland
has shown that termites are not dependent upon intestinal spirochaetes,
although he has not shown that the spirochartes do not play an im-
portant part in cellulose digestion.
In every experiment carried out in the present study with woody
material from termite nests, abundant growths of cellulose-digesting
organisms were obtained. The numerous cellulose-digesting organ-
isms, wrhich were undoubtedly taken into the digestive tract with food,
could not again be isolated from the intestine on the cellulose media
used.
The most conspicuous organisms, with reference to numbers and
bulk next to the Protozoa, as seen by microscopical examination of the
intestinal contents of termites, are the spirochaetes. Since these do not
grow on the usual laboratory media their true significance has not been
explained.
92 ALBERT DICKMAN
SUMMARY AND CONCLUSIONS
Cellulose-digesting organisms, both bacteria and molds, are very
abundant in termite nests. These organisms have been obtained in
abundance from material dried for a year.
Cellulose-digesting bacteria were not isolated from the intestines of
termites on the nitrate-cellulose medium of Bradley and Rettger nor
on the silica-gel medium of \Yinogradsky.
True bacteria are probably of little importance in cellulose digestion
in termites.
BIBLIOGRAPHY
BECKWITH, T. D., ROSE, EDYTHE J., 1929. Cellulose Digestion by Organisms from
the Termite Gut. Proc. Soc. Exper. Biol. and Med., 27: 4.
BRADLEY, L. A., RETTGER, L. F., 1927. Studies on Aerobic Bacteria Commonly
Concerned in the Decomposition of Cellulose. Jour. Bacterial., 13: 321.
BUCHNER, PAUL, 1928. Holznahrung und Symbiose. Berlin.
CLEVELAND, L. R., 1924. The Physiological and Symbiotic Relationships between
the Intestinal Protozoa of Termites and their Host, with Special Reference
to Reticulitermes llavipes Kollar. Biol. Bull., 46: 178.
Ci EVELAND, L. R., 1925. The Method by which Trichonympha campanula, a
Protozoon in the Intestine of Termites, Ingests Solid Particles of Wood for
Food. Biol. Bull., 48: 282.
CLEVELAND, L. R., 1925. The Ability of Termites to Live Perhaps Indefinitely on a
Diet of Pure Cellulose. Biol. Bull., 48: 289.
CLEVELAND, L. R., 1925. The Feeding Habit of Termite Castes and its Relation to
their Intestinal Flagellates. Biol. Bull., 48: 295.
CLEVELAND, L. R., 1925. The Effects of Oxygenation and Starvation on the Sym-
biosis between the Termite, Termopsis, and its Intestinal Flagellates.
Biol. Bull., 48: 309.
CLEVELAND, L. R., 1928. Further Observations and Experiments on the Symbiosis
between Termites and their Intestinal Protozoa. Biol. Bull., 54: 231.
DORS , \V. H., MILLER, R. C., 1923. The Digestion of Wood by Teredo navalis.
I'niv. Calif. Publ. in Zoo/., 22: 383.
HOELLING, B. A., 1910. Die Kernverhaltnisse von Fusiformis termitidis. Arch.
Protistenk., 19: 239.
HOLLANDE, A. C., 1922. Les Spirocht-tes des Termites; processus de division;
formation du Schizoplaste. Arch, de Zoo/. Exper. et Gen., 61: 23.
IMMS, A. D., 1924. A General Textbook of Entomology. London.
McBiiTH, I. G., 1916. Studies on the Decomposition of Cellulose in Soils. Soil
Science, 1: 437.
WAKSMAN, S. A., CAREY, C., 1926. The Use of the Silica Gel Plate for Demonstrating
the Occurrence and Abundance of Cellulose- Decomposing Bacteria. Jour.
Bacterial., 12: 87.
WINOGRADSKY, S., 1929. Etudes sur la microbiologie du sol. Ann. de I'Institut
Pasteur, 43 : 549.
THE INNERVATION OF THE STOMACH AND RECTUM
AND THE ACTION OF ADRENALINE IN
ELASMOBRANCH FISHES
BRENTON R. LUTZ
(From the Mount Desert Island Biological Laboratory, Maine and the Physiological
Laboratory of Boston University, School of Medicine)
A study of the literature concerning the innervation of the stomach
and intestine in mammals reveals much confusion and contradiction.
The orthodox differentiation into sympathetic and parasympathetic
with antagonistic actions has many exceptions. Thus Langley (1898)
found inhibitory fibers to the stomach in the vagus of the rabbit,
Morat (1893) found excitatory fibers to the stomach and intestine
in the splanchnic of the dog, and Carlson, Boyd and Pearcy (1922)
have found that both the splanchnics and the vagi of the cat carry
both kinds of fibers to the stomach. On the basis of effects produced
by adrenaline, Smith (1918) assumed the splanchnics to be inhibitory
for the stomach in man and in the cat, but only for certain parts of
the stomach in the guinea pig, rabbit and dog, while being motor for
other parts. Tashiro (1920), however, using adrenaline on surviving
cat intestine, came to the conclusion that there are motor fibers to the
circular muscle in the sympathetic nerves as well as inhibitory fibers
to both the circular and the longitudinal layers. McCrea, McSwiney
and Stopford (1925) found that in dogs, cats and rabbits the primary
effect of stimulation of the peripheral cut end of the vagus on the
stomach may be inhibition or augmentation, depending upon the
intragastric pressure, but that the final effect is motor. Brown,
McSwiney and Wadge (1930) found that the effect of sympathetic
stimulation depends on the type of stimulation. A low frequency
contracted the body of the stomach in the cat, whereas ordinary
tetanizing current inhibited. All rates inhibited the antrum, and
adrenaline caused an inhibition of both parts. In a review Van
Campenhout (1930) says, "We believe the actual distinction of
sympathetic, parasympathetic and local innervations to be erroneous
owing to ignorance of the real constitution of the visceral autonomic
nervous system." A similar view was expressed by Langfeldt (1929),
who concluded that there is no absolute antagonism between the
sympathetic and parasympathetic and that our information con-
cerning the peripheral termination of both systems is incomplete.
93
94 BREXTOX R. LITZ
The literature concerning the visceral innervation in the lower
vertebrates shows no more conformity to the orthodox view than does
that in mammals. Goltz (1872) showed that the splanchnic nerves
are motor for the stomach in the frog, and this has been confirmed by
Dixon (1902), Miiller and Liljestrand (1918) and Itagaki (1930).
1 >ixon (1902) found the vagus in the frog to have either an inhibitory
or a motor effect on the stomach. For reptiles there is not enough
evidence to make a comparison, either anatomical or physiological,
although Thorell (1927) by the use of adrenaline considered the
sympathetic to be inhibitory to all parts of the turtle's stomach
except the cardiac portion. In birds Xolf (1925) has reported that
the vagus is motor to the crop, and either motor or inhibitory to the
gizzard and small intestine; and the coeliac nerves are either motor
or inhibitory to the gizzard and duodenum.
In elasmobranch fishes the autonomic nervous system appears
not to be well developed (Miiller and Liljestrand, 1918). Bottazzi
(1902) found both the vagus and the anterior splanchnic nerves in
Scyllium canicula to be motor for the stomach. He was unable to
demonstrate any inhibitory effect of either. Stimulation of the cord
in the region from the forty-fifth to forty-eighth spinal nerves gave
motor activity of the rectum. Miiller and Liljestrand (1918) con-
firmed Botazzi (1902) in part, using Squalus acanthias and various
species of Rain, but believed an inhibitory effect of the vagus on the
stomach to be more marked than the motor effect. They never
obtained evidence of inhibition from the anterior splanchnics. Stimu-
lation of the middle and posterior splanchnic nerves was without
effect on the spiral valve and rectum.
In view of other peculiarities of the autonomic nervous system in
elasmobranchs, namely, the lack of accelerator nerves to the heart
(Bottazzi, 1902; Miiller and Liljestrand, 1918; Lut/., 1930</) and the
inhibitory action of adrenaline on the heart (Macdonald, 1925; Lutz,
19306), the present writer believed that it might prove useful to
compare the effects of adrenaline and extract of chromaphil tissue on
parts of the gut with the effects of electrical stimulation of the extrinsic
nerves to the same parts.
MATERIAL AND MKTIIOD
The elasmobranchs used were Squalus acanthias, Raia erinacea and
K. diaphanes. For anatomical reasons only S. acanthias was used
when nerves were stimulated. Segments of the stomach, spiral valve,
and rectum one half to one inch long were suspended in 50 cc. of a
physiological solution drscribrd by Lutz (19306), and tracings ob-
VISCERAL INNOVATION IN ELASMOBRANC I IS
95
tained. Some pieces were hung so that the circular muscles would
activate the lever, others were suspended so that the longitudinal
layer would be most effective. Adrenalin chloride (Parke, Davis &
Co.) and chromaphil tissue extracted in distilled water were added to
the bath by means of a pipette. Control tests, in which similar
amounts of distilled water and extracts of liver and spleen were added
to the bath, showed that the method was satisfactory. An extract
of the anterior chromaphil bodies was made in one cc. of distilled
water immediately on removal of the tissue, and used at once. In a
few cases, in which R. stabuliforis served as a source of chromaphil
tissue, only one "axillary heart" was used to one cc. of distilled water,
FIG. 1. Effect of adrenalin chloride, 1 in 50,000, on the pyloric portion of the
stomach. Time in 5 second intervals. A, Raia erinacea. Typical effect on tonus.
B, R. diaphanes. Effect mainly on motility.
but in the case of the smaller species of Raia both anterior chromaphil
bodies were used and sometimes, in addition, some of the accessory
bodies.
In the experiments in which the extrinsic nerves were stimulated
the entire central nervous system was pithed. The left vagus was
exposed through the anterior cardinal sinus. The first sympathetic
96
BRENTOX R. LUTZ
ganglion (gastric) and the anterior splanchnic nerves were exposed
through the posterior cardinal sinus, or the latter were sometimes
stimulated along the course of the coeliac artery. The posterior
splanchnic nerves (rectal) were stimulated along the posterior mesen-
FlG. 2. Effect of adrenalin chloride, 1 in 50,000, on the rectum of Squalus
acanthias. Time in 5 second intervals. At M, 2 cc. of the bath fluid squirted on the
preparation. Adrenalin chloride added at A.
teric artery or in the mesentery supporting the rectal gland. Records
of the movement of the gut were obtained by means of a small hook
and a light lever. The nerves were stimulated with faradic current
by means of platinum electrodes leading from an inductorium (Harvard
Apparatus Co.) with the secondary coil set at 8 cm. and one 2.5 volt
dry cell in the primary circuit.
RESULTS
Adrenalin chloride added to the bath fluid to make one in 50,000
caused a rise in tone and sometimes augmentation of rate and height
of the movements of the pylorus and other parts of the stomach in
twenty-five preparations .and had no effect in three cases (Fig. 1).
VISCERAL INNERVATION IN ELASMOBRANCHS
97
In some inactive preparations motility was initiated by a similar dose.
A distilled water extract of chromaphil tissue taken from the skate
gave the same effect as adrenalin chloride on the pylorus and stomach
of both the skate and the dogfish (Fig. 3). Extracts of liver and
spleen, agitation of the bath fluid, or the addition of distilled water
gave no response.
On twelve preparations of the posterior end of the spiral valve
and the rectum, adrenalin chloride, one in 50,000 caused a marked
fall in tone and an inhibition of motility (Fig. 2). In no case was
FIG. 3. Effect of extract of chromaphil tissue. Time in 5 second intervals.
A, Squalus acanthias, pyloric portion of the stomach. A', extract of the axillary bod-
ies from one side of Raid stabnliforis. B, R. erinacea, rectum. X, extract of the
axillary bodies of the same specimen.
there activation or failure of response. Extract of chromaphil tissue
also caused inhibition (Fig. 3). Extract of liver gave no response.
In thirteen specimens of 5. acanthias faradic stimulation of the
first sympathetic ganglion (gastric) or the anterior splanchnic nerves
caused extensive contractions of the stomach beginning in the pyloric
region. The latent period varied from two to five seconds (Fig. 4, A).
7
98
BRENTON R. LUTZ
In seven fishes faradic stimulation of the peripheral end of the
cut vagus at the anterior cardinal sinus caused moderate contractions
of the pylorus and adjacent region. The latent period was about five
seconds. In one case no response was obtained. The response from
the vagus was never obtained longer than forty minutes after the
IK.. 4. Kltert of faradic stimulation. Time in seconds. A , Squalus acanthias.
Contraction of the pyloric portion of the stomach on stimulating the first sympathetic
ganglion (gastric) for one half second. Latent period, 2.5 seconds. B, 5. acanthias.
Contraction of the rectum on stimulating the posterior splanchnic nerves for one
second. Latent period, 8 seconds.
opening of the cardinal sinuses, whereas the sympathetic response
was obtained after three hours.
In four fishes the posterior splanchnic nerves were stimulated and
in each ca>e a vigorous contraction of the rectum and the adjacent
VISCERAL INNERX ATION IN ELASMOBRANCHS 99
part of the spiral valve was repeatedly obtained (Fig. 4, B). The
response had a latent period of eight to ten seconds, and in one speci-
men was active after three hours without the circulation.
DISCUSSION
The motor effect of electrical stimulation of the sympathetic and
of adrenaline on the stomach of the elasmobranch is another exception
to the view that in general the sympathetic is inhibitory to the gut
and the vagus motor. In this case the effect of adrenaline is sym-
pathico-mimetic. If the posterior splanchnic nerves, stimulation of
which activates the rectum, are sympathetic, as Muller and Liljestrand
(1918) describe them to be, then the inhibitory action of adrenaline
on this part of the intestine is also an exception. Brown, McSwiney
and Wadge (1930) found that adrenaline did not reproduce the
effects of sympathetic stimulation of the stomach in the cat and in the
dog, inhibition being the invariable result.
The results reported here confirm Bottazzi (1902) working on
Scyllium, and Muller and Liljestrand (1918) working on Squalus and
Raid insofar as the effect on the stomach of electrical stimulation of
the anterior splanchnic nerves is concerned. However, since a marked
contraction of the rectum resulted from stimulation of the posterior
splanchnic nerves, and no evidence of inhibition of the stomach
through stimulation of the vagus was obtained, these results are to
that extent at variance with those of Muller and Liljestrand.
While there may be a valid reason for perpetuating the morpho-
logical division of the autonomic nervous system into cranial, thoraco-
lumbar and sacral parts, there is sufficient evidence to indicate that
a general physiological distinction should not be made so far as control
of the alimentary tract is concerned.
SUMMARY
1. Adrenalin chloride and extract of "the chromaphil bodies caused
a rise in tone and sometimes an increase in motility of all parts of the
stomach of Squalus acanthias, Raia erinacea and R. diaphanes.
2. Faradic stimulation of the first sympathetic ganglion (gastric)
and the anterior splanchnic nerves caused extensive contractions of
the stomach beginning at the pylorus in Squalus acanthias. Similar
stimulation of the vagus caused moderate activity in the region of
the pylorus.
3. Adrenalin chloride and extract of chromaphil bodies caused a
marked decrease in tone and inhibition of motility of the posterior
end of the spiral valve and the rectum in all three elasmobranchs.
100 BREXTOX R. LUTZ
4. Faradic stimulation of the posterior splanchnic nerves caused a
vigorous contraction of the rectum and adjacent part of the spiral
valve in Sqnalus acanlhias.
5. The data presented here and the evidence from the literature
indicate that a general physiological distinction between the sympa-
thetic and the parasympathetic divisions of the autonomic nervous
system should not be made.
BIBLIOGRAPHY
BOTTAZZI, F.( 1902. Zeitschr.f. Bio!., 43: 372.
BROWN, G. L., McSwixEY, B. A., AXD WADGE, \V. J., 1930. Jour. Physiol., 70: 253.
CARLSON, A. J.? BOYD, T. E., AND PEARCV, J. F., 1922. Am. Jour. Physiol., 61: 14.
DIXON, VV. E., 1902. Jour. Physiol., 28: 57.
GOLTZ, F. 1872. Pfliiger's Arch., 6: 616.
ITAGAKI, M., 1930. Jap. Jour. Med. ScL, 1: 105.
LAXGFELT, G., 1929. En oversikt over den kliniske under sokelse av det viscerale
nervensystem og en kritikk de forskjellige pr overs praktiske verdi. Bergen.
LAXGLEY, J. X., 1898. Jour. Physiol., 23: 407.
LUTZ, B. R., 1930a. Biol. Bull., 59: 211.
LUTZ, B. R. 19306. .4m. Jour. Physiol., 94: 135.
MACDONALD, A. D., 1925. Quart. Jour. Exper. Physiol., 15: 69.
McCRHA, E. D., McSwiNEY, B. A., AXD STOPFORD, J. S. B., 1925. Quart. Jour.
Exper. Physiol., 15: 201.
MORAT, J. P., 1893. Arch, de physiol. norm, el path., 25: 142.
•MULLER, E., AXD LiLjESTRAND, G., 1918. Arch. Anat. u. Physiol., Anat. Abt., p. 137.
NOLF, P., 1925. Arch, internal, de physiol., 25: 291.
SMITH, M. I., 1918. Am. Jour. Physiol., 46: 232.
TASHIRO, K., 1920. Tohoku Jour. Exper. Med., 1: 102.
THOREI.L, G., 1927. Skand. Arch. f. Physiol., 50: 205.
VAN CAMPENHOLT, E., 1930. Quart. Rev. Biol., 5: 217.
LABORATORY REPRODUCTION STUDIES ON THE GROUND
SQUIRREL, CITELLUS TRIDECEMLINEATUS
PALLID US, ALLEN1
GEORGE E. JOHNSON AXD NELSON J. WADE
KANSAS STATE AGRICULTURAL EXPERIMENT STATION
INTRODUCTION
In studies on hibernation carried on in this laboratory for the past
six years hundreds of ground squirrels, Citellus tridecemlineatus , have
been kept in our animal house. During this time many pregnant
females have been received in the spring and have usually reared their
young, but mating has been known to occur in the laboratory only in
two females, both C. t. pallidus Allen,2 the variety used in these ex-
periments. Since the animals were well cared for and were in good
health, it seemed desirable to investigate the possible causes of their
sterility for the scientific as well as the practical information which
such a study might yield. It was expected that reproduction was un-
likely to occur except at about the time of the normal breeding season
in April and May (Drips, 1919; O. Wade, 1927), but why rut is limited
to the spring in this and many other species is another question which
studies of the present type may in time help answer.
For the sake of brevity the specific methods and the literature will
be considered with the different types of experiments. The majority
of the animals in the laboratory served as controls for the special
experiments. These controls were kept in wire cages with wood
bottoms measuring two by three feet. From one pair to about six
animals were usually kept in one cage. Wood shavings were used on
the floor and small wooden boxes in which the animals could build
nests were usually provided.
OBSERVATIONS ON CONTROLS
Since our ultra-violet light, outdoor cage and ovarian extract ex-
periments were performed between March and June, 1930, inclusive,
a group of 6 males and 11 females were observed as special controls
1 Contribution No. 132 from the Department of Zoology, Kansas State Agri-
cultural College, Manhattan.
2 As these animals were secured in central western Kansas it is possible that some
of them may have been C. t. arenicola the southern variety recently split off from C. t.
pallidus by Howell (Proc. Biol. Soc. Wash., 41: 213, 1928).
101
102 G. E. JOHNSON AND N. J. \VADE
during these UK tilths. In three of these males the testes were partly
enlarged and had migrated from the abdomen, the position during the
fall and early winter, partly into the scrotum which was enlarging.
The latter condition of enlargement of the testes and descent into the
-rroumi, which becomes darkly pigmented, is typical of the breeding
season and animals in this condition will be referred to as "scrotal."
Animals showing partial enlargement of the testes and partial descent
wit In nit enlargement of the scrotum will be called "partly scrotal."
In the eleven control females only one had the vagina open between
March and June.
As the pituitary implantation experiments and one ovarian extract
experiment were not limited to tin- spring, it is important to refer to
tin- sexual development over a longer period than that given for the
special controls already mentioned. For years it has been observed
that captive male ground squirrels became "scrotal " in late winter and
in the spring, some being in this condition when others were not.
Several females had also been seen with swollen and even open vaginae.
\Yhile the observations from March to June, 1930 showed only a small
proportion of animals with enlarged external genitalia, probably be-
cause of the lateness, observations made in 1931 with the assistance
of Mark A. Foster showed a pronounced development in January and
February. In fact a majority of males were "scrotal" January 10.
The females, with one exception, showed no external swelling on this
date. By February 1 all of the healthy males (26) were scrotal and
nine of twelve females more than one year old had very swollen
vagina-, and three of the nine had the vagina open. Of the females
which were less than one year old, seven of sixteen had very swollen
bill not open vagina-.
Since males and females have nearly always been together in our
laboratory there should have been ample opportunity for a great many
to breed each year, but none of the animals which have been in the
laboratory over winter ha\e \ct been known to breed, and only two
cases of reproduction among newly received ground squirrels are on
record in this laboratory. One female received May 1, 1930 gave birth
to a litter on June 22, 1930, and therefore must have bred in the
laboratory about May 25. In 1926 one case of breeding in the labora-
tory had been witnessed on April 24, the day the animals were received.
The litter was born about 27.5 days later. This is in agreement with
data for ('. /. tridecemlineatus by Drips (1919), who gives the period of
gestation as 28 days, and by O. XYade (1927), who reports two cases of
reproduction in the laboratory in which it was between 27 and 28 days.
REPRODUCTION STUDIES ON GROUND SQUIRREL 103
NUTRITION AND VITAMINS
The animals were fed a ration which we had found to maintain good
health and reproduction in rats and mice. It consisted of yellow corn
meal (30 per cent), whole wheat flour (30 per cent), skim milk powder
(30 per cent), alfalfa meal (4 per cent), bone meal (4 per cent), salt
( 1 per cent), and cod liver oil (1 per cent). The cod liver oil was added
just before feeding. Sprouted oats or green feed and water were also
supplied.
The well nourished appearance of the animals showed this diet to be
adequate for health. Vitamin A, which may have a slight influence on
reproduction (Sure, 1928; Evans, 1928«), was present in the cod liver
oil, wheat and corn. Vitamin B, whose absence might produce a poor
physical state with loss of reproductive power in the male (Evans,
19286; Mattill, 1927) or cessation of oestrus in the female in four weeks,
with death some two months later (Parkes, 1928), was supplied by
the wheat, corn and alfalfa. Vitamins C and D apparently have little
effect on reproduction but they were supplied, C by the green feed and
D by the cod liver oil, wheat and green feed. An absence of vitamin E
would produce a degeneration of seminal epithelium in the male and
an early resorption of the young in the female (Evans, 1925; Evans,
Burr and Althausen, 1927), but this vitamin was supplied by the
wheat, corn, alfalfa and green feed. The addition of wheat "germ
stock" which is especially rich in vitamin E, to the diet of 24 animals
through the month of June, which would be the latter part of the nor-
mal breeding season, did not affect genital development or reproduc-
tion.
PITUITARY IMPLANTS
The work of Smith (1927«, &) and Smith and Engle (1927) has
shown that implantation of the anterior lobe of the pituitary may
produce precocious sexual maturity and super-ovulation in mice and
rats. These authors have shown that similar results are obtained from
implants of the whole gland, indicating that the inclusion of the pos-
terior lobe does not affect the action of the anterior lobe in such im-
plants. In attempting to stimulate the reproductive organs of the
ground squirrels, implants of pituitary glands from rats were made into
adult ground squirrels. While the entire gland was used, the results
obtained should be attributed to the anterior lobe alone as already
indicated. About thirty animals were used from November to August.
Those used during the summer had been in the laboratory over winter
and therefore were at least one year old. A finely cut pituitary from a
rat was implanted or injected, with a small amount of physiological salt
104 G. E. JOHNSON AND N. J. WADE
solution, subcutaneously in the hind leg of the anaesthetized animal by
means of a fine glass canula. A striking external genital enlargement
was observed in three days, i.e., as a result of two daily implants, in
most of the animals. In the female this consisted of a swelling of the
vulva and was usually followed by the opening of the vagina after
three to six daily implants. In the male the testes usually showed
some enlargement and tendency to become scrotal after two implants,
and successive implants usually produced still greater enlargement and
migration into the scrotum, which usually became more pigmented.
The changes in the male were not so pronounced as in the female, for
the testes rarely became as large as normal if the implants were made at
other times than the breeding season. These genital changes usually
persisted about two weeks or more, but they resulted in no cases of
reproduction.
Daily implantation of a rat pituitary into each of 8 male and 8
female ground squirrels was begun June 7. By killing one pair on
June 8 and each day thereafter a series was obtained, one pair having
received one implant each, another pair, two implants, etc., up to a
pair which had received eight daily implants each. The ovaries of the
females showed a general increase in size, although not showing a
perfectly graded series from the one to eight implants, probably be-
cause of differences in the animals before the implants were made.
A control female, which had not received any implants, and also the
experimental female which had received three implants had corpora
lutea apparently of oestrus, since no indications of pregnancy were
found. This implanted animal therefore did not fit into the series.
The ovary of the animal which had received only one implant of one
rat pituitary gland contained: numerous small oogonia, without sur-
rounding follicular cells, near the periphery of the ovary; numerous
small atr<.-ti< follicles; and a few moderately large follicles. The ani-
mals which had received two and four daily implants showed: an in-
crease in size and number of the larger follicles; some reduction in
number of atretic follicles; and possibly some reduction in the number
of peripheral oogonia. Xo corpora lutea were yet present, but a few of
the larger follicles in the animal receiving two implants contained
blood. The chief change in the animals receiving 5, 7 and 8 daily
implants was the tendency of the large follicles to become corpora
lutea, and this included even the ones filled with blood. The atretic
follicles and also the peripheral oogonia remained about the same in
number and size or were possibly reduced in number. Apparently the
implants had little effect on the oogonia and the atretic follicles in eight
days, but produced a striking increase in size of large or medium sized
REPRODUCTION STUDIES ON GROUND SQUIRREL
105
PLAT,: 1!
Ib
3
FIG. 1. a. A section through a portion of the right ovary of ground squirrel
Xo. 1011 before receiving implants, X 23. b. A portion of the same section, X 83.
The large follicles measure about 340 microns in diameter.
For further description of the figures see the text.
All the photomicrographs were taken by Mr. Charles Dobrovolny.
FIG. 2. A section through the remainder of the right ovary of ground squirrel
No. 1011 after receiving 6 daily implants of rat pituitaries, X 23. The larger follicles
measure about one mm. in their greatest diameter.
FIG. 3. A section through the left ovary of ground squirrel No. 1011 after 8
daily implants of rat pituitaries, X 23. The larger follicles now measure about 1.2
to 1.5 mm. in their greatest diameter.
106 G. E. JOHNSON AND X. J. WADE
follicles, many of which were hemorrhagic. and finally their develop-
ment into corpora lutca.
More Mriking results were obtained from histological study of the
ovaries <>t~ two female^ which served as their own controls. On August
8 half of tin- right ovary of each was removed for study of their normal
condition. Each received an implant of a rat pituitary daily from
August 13 to 21. The remainder of the right ovary was removed on
the seventh day (after 6 pituitary implants) and the left ovary taken
on August 22, the ninth day (after 8 daily implants). The normal or
control piece of the right ovary of one animal contained a few oogonia
without follicles, some atretic follicles and a few young to medium-
sized follicles (Fig. 1). After 6 days of implantation the follicles in-
creased in size and some of them were then mature (Fig. 2); after 8
implants these were extremely large, much larger than normal mature
follicles, and some contained red blood cells in a part of the follicular
cavity (Fig. 3). No marked change in number of atretic follicles and
of oogonia could be noted.
The control piece of ovary removed from the other animal, August
8. contained main- oogonia without follicles, very many atretic small
follicles, some small follicles and some which had grown to the stage of
beginning of cavity formation. Six days of implantation produced a
few large follicles, most of which contained blood. After eight days
the follicles were still larger, some of them containing blood, and cor-
pora lutea were beginning to form, even beginning to hem in the blood
in some of the follicles. Two ova were seen in one fallopian tube of
this animal. In this and also in the other female killed August 22,
one day after each had received eight daily implants of rat pituitaries,
the ovaries and uteri were much enlarged and the vagina- were swollen
and open at the time of autopsy.
The males killed, one each day for 8 days, after one to eight days of
pituitary implantation beginning June 8 did not show a perfectly
graded series as to histological developnu-nt ot the testes, possibly
because of variations in condition at the beginning of the experiment.
A non-implanted male, killed at tln^ time as a control, showed many
more spermatogonia and more spermatogonial divisions than the im-
planted males, and many primary spermatocyte (spireme) stages, but
without more advanced stages of mitosis. The animals implanted one
and two days showed little advance over the control. Those implanted
three and four da\> >howed testicular enlargement and some descent
of the testes into the scrotum, and sections of the testes showed chiefly
>|)fi-inaii<U and attached >permatoxoa in addition to spermatogonia.
Metamorphosing spermatids were seen in the four-day implanted ani-
REPRODUCTION STUDIES OX GROUND SQUIRREL 107
mnl. In these two males implanted for three and four days, the pros-
tates and Cowper's glands were large, but this was not the case in any
of the other 6 males. Probably these two males were more sexually
active before the implantations than the others were, otherwise these
two should not have been the only ones to show enlargement of these
glands. The last four males implanted showed enlargement of the
scrotum and some enlargement of the testes but not as marked as in
the ones implanted three and four days. A gradual histological change
occurred in these four males from a predominance of primary sperma-
tocyte spiremes (after 5 days of implantation) through some stages of
active mitosis of the first maturation division (6 and 7 days), to stages
of active mitosis of the second maturation division (8 days). Sperma-
togonia were present, but spermatids and spermatozoa were not found
in these four animals.
From each of two males, which had been in the animal house a year,
a small piece of testis was removed for study of the testicular condition
before pituitary implantation, August 9. These control pieces con-
tained spermatogonia and many primary spermatocyte spireme stages
without more advanced mitotic stages (Fig. 4). After four days of
healing these two males were given daily implants of rat pituitaries for
8 days and the testes fixed on the ninth day, August 21. The testes of
one of the males now contained many spermatids and active division
phases of mitosis of maturation (Fig. 5). The testes of the other male
now showed a few spermatids and several active division phases of
mitosis of maturation (Fig. 6). No metamorphosing spermatids and
no spermatozoa were present in either male.
While the development in the gonads was not uniformly progressive
in the eight-day series of animals, it is apparent that the implants in
the female generally stimulated the growth of the follicles to an unusual
size, often accompanied by bleeding into the follicle; and in the male
produced a general enlargement of the testes tending towards the form-
ation of spermatids, but without enlargement of the prostate and
Cowper's glands to the extent found in breeding males.
Work on the conditions of the reproductive organs at different
seasons, and on the effect of pituitary implantation at different times
of the year on our laboratory supply of ground squirrels, is being con-
tinued by the senior author and Mr. Mark A. Foster.
ULTRA-VIOLET RADIATIONS
Saidman (1924) found ultra-violet light beneficial in treating men-
strual disturbances. At the suggestion of Dr. R. K. Nabours of this
department a mercury arc ultra-violet lamp was placed at a distance
108 G. E. JOHNSON AND N. J. WADE
of three to four feet above several all-wire cages containing ground
squirrels. The lamp transmits slightly more ultra-violet light than is
found in sunlight, and will cure rickets in chickens in half-hour daily
radiations according to Professor J. S. Hughes of the Chemistry De-
partment of this College. However, radiations of five to twenty-four
hours daily had no influence on the genital or general physiological
condition of our ground squirrels.
A combination of ultra-violet radiation with pituitary implantation
produced no effects other than those of implantation alone.
OVARIAN EXTRACT INJECTIONS
(Estrus was produced in old albino rats by Slonaker (1927) and in
castrated and normal mice by Tuisk (1927) with follicular fluid ex-
tracts. Golding and Ramirez (1928), by the use of ovarian and pla-
cental extracts, caused the vagina? of rats to open prematurely with a
production of continuous oestrus. \York by Allen and Doisy (1923)
and others a^so show that ovarian extracts may have a marked effect
upon genital changes.
Through the courtesy of the Veterinary Division of this College we
procured an alcoholic saline extract of beef ovaries from which the
corpora lutea had been removed. This extract had been used with
success upon non-producing cows by McLeod (1929) and Frank (1929).
Injections of one cc. of this extract every third day for the last two
weeks in June produced no positive changes in female ground squirrels
either with the vagina open or closed, although the dosage was about
200 times as great as that used for cows when relative weights are con-
sidered. The females with open vagina? were placed daily with sexually
active males, but no reproduction took place.
Another experiment was begun January 14, 1931. The extract was
injected into three female ground squirrels, one cc. daily for 22 days.
It produced no apparent effect on the vagina, which was closed and not
swollen at the beginning of the experiment. Histological examination
of one-third to one-half the right ovary taken from each animal and
from one control January 14 showed a large number of young oogonia
(primitive ova) peripherally, many young follicles with one row of
follicular cells, several small atretic follicles, and a medium number of
mature follicles about 300 to 360 microns in their greatest diameter.
FH;. 4. A section through a portion of the right testis of ground squirrel No.
1115 before it received any pituitary implants, X 120.
Fi<;. 5. A section through the remainder of the right testis of ground squirrel
No. 1115 after receiving 8 daily implants of rat pituitaries, X 120.
FIG. 6. A section through a testis of ground squirrel No. 1078 after 8-daily ini-
pl. nits of rat pituitaries, X 120.
REPRODUCTION STUDIES ON GROUND SOI IRREL
109
'.
'* " 0 '"" *.
*• -, •" *
."•C: . i
110 G. E. JOHXSOX AM) X. J. WADE
Careful histological study of the remaining half or two-thirds of the
ovary after 13 daily injections showed no marked or consistent change
in the two experimentals surviving and in the saline injected control.
Xo histological change was found after 21 days of injection in the one
surviving experimental and in the control. In both animals the left
ovary was slightly larger than the right had been at the beginning of
the experiment, and the uterus had increased about fifty per cent in
diameter. Both of these conditions may probably be attributed to the
approach of the spring breeding season. The health of these animals
was good, the two deaths being produced by the animal chewing into
the incision or the somewhat inflamed areas where the injections had
been made.
• OVARIAN IMPLANTATION
Ovarian transplantation in the hands of other workers has had
marked effect upon the recipients. Grunert (1927) produced oestrus
and pregnancy in cows with homeotransplants. Sippel (1924) re-
ported pregnancy in four women after ovarian grafts, and Turner and
Bour (1925) reported improved health and return of menses with the
possibility of pregnancy after such implants. Pettinari (1925) re-
activated an old female dog with ovarian grafts.
These and other reports suggested the possibility that implants of
ovarian ti>Mie might stimulate genital functions. It was not expected
that the tissue would grow, partly because rat ovaries were used, but
that it would release the contained hormone as the tissue was gradually
absorbed, as wras the case with the pituitary ti>sue implants. The
method and technic were practically identical in the two cases. The
animals were implanted every third day for two weeks.
The four females used showred no genital development or changes.
These implants were not tolerated as well as the pituitary implants,
but no serious ill effects were produced.
OUTDOOR CAGKS
Since taking ground squirrels from their native habitat into the
l.ibor.itory stopped their reproduction, it would seem that they should
reproduce if placed in outdoor cages. Six pairs kept in such cages
during the month of June showed no genital changes when dug out of
their burrows. Whether reproduction would take place if the animals
were in outdoor cages in April when most of the mating takes place is
not known. Hven if they did, this would not show why reproduction
did not occur in the laboratory.
REPRODUCTION STUDIES ON GROUND SQUIRREL 111
DISCUSSION
The Possible Inhibitors of Reproduction in the Laboratory
Since the use of the mercury arc vapor lamp did not aid reproduc-
tion, the lack of sunlight would not seem to be an inhibitor of it. The
benefits of the outdoor sun should last for some time after the animal
is captured. Furthermore, it may be stated that the Columbian
ground squirrels mate early in the spring before they have had time to
absorb much ultra-violet light (Shaw, 1926). This would be true to
some extent for the thirteen-lined ground squirrels, for they have a
period of rut of about two weeks, shortly after coming out of hiberna-
tion, according to Drips (1919).
As the various vitamins were provided in the diet, lack of reproduc-
tion could not be attributed to their absence. The good condition of
the animals, the good growth made by young ground squirrels, and also
the satisfactory rate of growth and reproduction in the mice and rats
fed the same diet give further evidence in that direction.
It cannot be stated that hibernation is not a necessary forerunner of
reproduction, but at least it does not greatly aid reproduction in the
laboratory, for great numbers of ground squirrels have been received
before and during the normal breeding season (April and May), but
only two known cases of copulation have occurred after they reached
the laboratory, although the females were nearly always left with
males. Such animals had, of course, passed through normal hiberna-
tion in nature. Ground squirrels which had hibernated in the refrig-
erator for varying lengths of time never reproduced following their
return to the animal house.
While marked development of the reproductive organs was pro-
duced in the ground squirrel by the anterior lobe principle in the pitui-
tary implants, this development was not as complete as that obtained
by Smith and his associates in the rat and mouse, or by Wolf (1929) in
the frog. Ovulation did occur in one of the implanted ground squirrels
in August. Possibly the female wrould have reproduced at this time if
the males had not been devoid of spermatozoa. Incidentally, the
ovulation in this female suggests that under these conditions at least,
ovulation may take place independently of copulation. Drips (1919)
stated that ovulation occurs only after coitus as found for the ferret
(Marshall, 1922) and rabbit (Hammond and Marshall, 1925). She
also stated that corpora lutea persist throughout the summer and
prevent the recurrence of oestrus after one litter is born.
Indications were found both in the literature and in our observa-
tions that the nervous state of an animal may have a profound influ-
112 G. E. JOHNSON AND N. J. WADE
ence on reproduction. Borries (1929) reported menstrual irregularities
in 27 per cent of 39 college women because of nervous strain. Lafora
(1923) considered that sexual frigidity in women was often produced
by psychic inhibition. According to Steive (1926, 1927) confinement
usually results in sterility and gonadal degeneration in wild animals.
Testicular atrophy has been noted in men in prison and war amenorrhea
in women has been produced by anxiety and worry. Sterility in
women was attributed to nervous strain by Macomber (1924). Dr.
Voder of the State Hospital, Kalamazoo, Michigan, in a letter states
that there is a positive correlation between genital functional derange-
ment and insanity.
Observations made on wild rats, captured to supply pituitaries for
the implant work, have indicated that nervous conditions produced
by confinement tended to inhibit reproduction. These rats would tight
so violently that only one or two could be kept safely in a cage. A
number of these, where physical injury was eliminated, died from no
apparent cause other than nervousness. A few of the tamest of these
wild rats were kept for several months in cages, some with wild, and
some with white rats, of the opposite sex, but without producing. The
two strains will mate, however, for it was found that an escaped albino
had mated with a gray male before she was captured again. A gray
female, pregnant when caught, gave birth to 6 young but killed them.
The ground squirrels with which we are working never become tame
in our laboratory. Gloves are practically a necessity in handling them,
since they resist capture and bite freely when cornered. It is interest-
ing to note that those handled most bite the most and those handled
least are the least wild. One animal received in the spring was un-
usually tame but in spite of frequent handling became as wild as the
others in a few weeks.
In reviewing our experimental work, we find that all of the methods
used to stimulate reproductive activity, including those found valuable
in other animals, were largely unsuccessful in the ground squirrels.
The ineffectiveness of these methods together with the nervous condi-
tion of the animals, taken with the influence of the nervous system on
genital function in other animals, suggests strongly that the almost
complete failure of the ground squirrels to reproduce in the laboratory
can be attributed to a nervous inhibition. If this is true, the nervous
condition probably acts indirectly through the endocrine system.
It is, furthermore, uncertain if larger doses of the hormones of the
proper endocrine organs can override this inhibition. The cause of
these abnormal genital conditions may lie in the nervous and endo-
crine systems working together to interfere with one or more steps in
normal reproductive process.
REPRODUCTION STUDIES OX GROUND SQUIRREL
SUMMARY AND CONCLUSIONS
113
1. The ground squirrel, Citellns tridecemlineatns, failed to reproduce
in nearly all cases under usual laboratory conditions.
2. The diet provided the animals contained all the vitamins which
have been found to be necessary for reproduction.
3. Pituitary implants from rats, with or without ultra-violet radia-
tion, did not cause reproduction but did stimulate the uterus and
follicles in the ovary to excessive growth and corpus luteum formation.
In the male the normal reproductive condition was not obtained at
other times than the breeding season, but the implants produced a
marked change in the histology of the testes from spireme stages of
primary spermatocytes to spermatocyte division stages and spermatids.
4. Ovarian implants, ovarian extract injections, ultra-violet radia-
tions, and keeping animals in outdoor cages caused no apparent effect
on the genital functions.
5. The cause of the inhibition of reproduction and of the failure to
reactivate the animals by various means may lie in a complex linkage
of the endocrine and nervous systems.
LITERATURE CITED
ALLEN, E., AND DOISY, E. A., 1923. An Ovarian Hormone. Jour. Am. Med.
Assn., Chicago, 81: 819.
BELLERBY, C. \V., 1928. Relation of Anterior Lobe of Pituitary to Reproductive
Organs. Lancet., 1: 1168.
BORRIES, KARA VON, 1929. Zur Frage der biologischen Wirkungen des Frauen-
studiums. Arch. f. Rassen- u. Gesellsch.-BioL, 22: 51.
DRIPS, DELLA, 1919. Studies on the Ovary of the Spermophile with Special Refer-
ence to the Corpus Luteum. Am. Jour. Anat., 25: 117.
EVANS, H. M., 1925. The Anti-sterility Vitamine Fat Soluble E. Science, 61: 519.
EVANS, H. M., 1928a. Effects of Inadequate Vitamin A on Sexual Physiology of
Females. Jour. Biol. Chem., 77: 651.
EVANS, H. M., 19286. Effect of Inadequate Vitamin B upon Sexual Physiology in
the Male. Jour. Nutrition, 1: 1.
EVANS, H. M., BURR, G. O., AND ALTHAUSEN, T. L., 1927. The Anti-sterility
Vitamine Fat Soluble E. Mem. Univ. Calif., 8:1.
FRANK, E. R., 1929. Observations on the Use of Ovarian Extract in the Treatment of
Sterility in Cattle. Cornell Veler., 19: 399.
GOLDING, GEORGE T., AND RAMIREZ, F. T., 1928. Ovarian and Placenta! Hormone
Effects in Xormal, Immature Albino Rats. Endocrinology, 12: 804.
GRUNERT. C. H., 1927. Absence of Estrum Corrected by Ovary Transplantation.
Vet. Med., 22: 112.
HAMMOND, J., AND MARSHALI , F. H. A., 1925. Reproduction in the Rabbit. Edin-
burgh.
LAFoRA, G. R., 1923. Sexual Frigidity in Women. Siglo med., 72: 105.:
MACOMBER, DONALD, 1924. Prevention of Sterility. Jour. Am. Med. Assn.,
Chicago, 83: 678.
MARSHALL, F. H. A., 1922. The Physiology of Reproduction. London.
McLEOD, W. M., 1929. The Use of Ovarian Extracts in Treatment of Sterility.
Cornell Veter., 19: 401.
114 G. E. JOHNSON AND N. J. WADE
MATTILL, H. A., 1927. The Relation of Vitamins B and E to Fertility in the Male
Rat. Am. Jour. PhysioL, 79: 305.
PARKED, A. S., l''2S. The Nature of the Anoestrous Condition Resulting from Vita-
min B Deficiency. Quart. Jour. Exper. PhysioL, 18: 397.
PETTI.VAKI. \ ., 1025. Phenornenes regeneratifs clans les ovaires d'une vielle chienne
apres greffe ovarienne. Compt. rend. Sac. de Bio/., 92: 1294.
OMAN, J.. 1924. Note sur les rayons ultra-violets et le traitement des glandes a
-< Vretion interne (ovaires). Bull Acad. de Mcd., 92: 938.
Mi\\v, \\ \i. T., 1926. A Short Season and Its Effect upon the Preparation for
Reproduction by the Columbian dround Squirrel. Ecology, 7: 136.
siri'ii, P.. 1924. Schwangerschaft nach homoioplastischer Ovarientransplantation
bei Hypovarismus. Zentralbl. f. Gytiak., 48: 15.
• '\AKER, J. R., 1927. The Effect of the Foliicular Hormone on Old Albino Rat-.
Am. Jour. PhysioL, 81: 325.
SMITH, P. E., 1927a. The Indu tion of Precocious Sexual Maturity by Pituitary
Homeotransplants. Am. Jour. PhysioL, 80: 114.
SMITH, P. E., l<;27/>. (ienital Sv-ieni Re^><>nM> iii Daily, Pituitary Transplants.
. . So, . Kxper. Bi»l. and Med., 24: 337.
SMITH, P. E., AND EXGLE, E. T., 1927. Induction of Precocious Sexual Maturity in
Mouse by Daily Pituitary 1 lomeo and 1 leterotransplants. Proc. Sac. Exper.
Biol. and Med.. 24: 561.
STIKVK, II.. 1('26. I'nfruchtbarkeit als Kolge unnatiirlichrr I,cbens\veise. (irenx-
fiagen des Nerven und SccL-nlebens, 126: 52. J. Bergmann, Munich.
SIIHVK, II., l')27. Die Abhiingigkeit der Keimdriisen vom Zustand des C'lesanit-
ki rpers und von dt-r I'mgebung. Xaturuiss., 15: 951.
SURE, BARXETT, 1('2S. Dietary Requirements for Fertility and Lactation: The
Vitamin A Content of Wheat Oil. Jour. A^r. Res. 37:93.
Ti i i II:K, Til., AND P.DIR, 1)., 1925. Menstruation and Pregnancy after Ovary
('.rafting or Transposition. Presse lu/'ti.. 33: 1073.
TuiSK, ROBERT, 1(>_'7. Protracted (Estrus Induced by Ovarian Extracts. Jour.
PhysioL 63: 180.
\\ \hi , Ons, 1927. Breeding Habits and Early Life History of the Thirteen-Striped
('.round Squirrel, Cildlux trideamliiu-iitiix (Mitchill). Jour. Mammal., 8:
269.
WOLF, OPAL M., 1929. ICffect of Daily Transplants of Anterior Lobe of Pituitary on
Reproduction of Frog \Rana pi/nrii* Shreberi. Proc. Soc. Exper. Biol. and
Mcd.. 26: 692.
A NEW PENTAMEROUS HYDROMEDUSA FROM
THE TORTUGAS
M. D. BURKENROAD
(From the Department of Zoology, Tnlanc University, and the Bureau of Research,
Department of Con serration of Louisiana)
During the month of July, 1929, a radially symmetrical pentame-
rous leptomedusa was present in very large numbers in the waters off
the Dry Tortugas, Florida. This medusa was quite similar to Pseudo-
clytia pentata, described by Dr. A. G. Mayer from the same waters;
in fact, it was identical with P. pentata in general form, in color,
in habits, and in time of appearance. It differed from P. pentata,
however, in certain important and specific structural characters, to be
described below.
Mayer (1910) says of P. pentata that it was "exceedingly abundant
at Tortugas, Florida, from June to August, 1897 to 1904. In 1905 it
was relatively rare . . . and not a single specimen could be found in
1908. ... In 1909 it again appeared in fair numbers." The animal
was thus abundant, in its season, for seven successive years, then rare
or not present for four successive years, reappearing in the next year.
Xo extensive recorded towings were made at the Tortugas from 1909
to 1929. Upon the reinvestigation of the pelagic fauna in 1929
(Grave and Burkenroad, 1929), a pentamerous medusa was found to
be present during July in such numbers as to be the dominant plankton
form, but this medusa was found to be not specifically identical with
the form described by Mayer. Pseudodytia pentata was not taken
at any time during the Laboratory season, from June 1 to August 18,
1929, while the medusa taken in 1929 had not been found by Mayer
in the course of many years of intensive investigation. The fact that
so similar a form appeared at the time when P. pentata might have
been expected to appear suggests the rapid and complete replacement
of one species by a related one, perhaps in the usual manner of invasion
by a species whose range is thus extended, but perhaps — and this
seems a most interesting possibility — through the complete replace-
ment of a parent stock by a successful, recent, and local mutant from
that stock. Mayer considered his Psendoclylia as probably derived
by mutation from some four-rayed C/v/w-like ancestor. Both in
1899 and 1909 he found P. pentata to be extremely variable, although
the variations do not appear to have been in the direction of the
115
116
M. D. BURKENROAD
medusa which was present in 1929. This last-found form was also
extremely variable, and its aberrations appeared to be in the same
direction and manner as those of P. pentata. The description of the
medusa which was present in 1929 follows:
Pseudoclytia longleyi,1 n. sp.
Adult Medusa. — Bell flatter than a hemisphere, 4 to 8 mm. in
diameter. Gelatinous substance of the bell thin and tenuous. Fifteen
to 20 simple tentacles with fairly well-developed, roundly conical
basal bulbs. Tentacles, when extended, twice the bell radius or more
in length; often carried contracted in close helical coils. Numerous
shaftless permanently rudimentary tentacle bulbs which vary in degree
of development from slight swellings of the ring-canal to bulbs as
FIG. 1. Pseudoclytia longleyi, n. sp. Diagrammatic. Typical arrangement of
the tentacles (T), rudimentary bulbs (B), and lithocysts (L). Tentacle shafts not
shown full-length.
large and as well-defined as those bearing shafts; from one to five
bulbs, usually two or three, between cadi pair of tentacles. One to
three lithocysts, usually one, between each pair of tentacles and
tentacle bulbs, so that the usual total number of lithocysts is about
fifty. Each lithocyst contains a single spherical concretion. Velum
well-developed, wide. There are five straight, narrow radial canals,
72 degrees apart. The five short, small gonads, of oval outline in
the male, circular in the female, are situated upon the radial canals
at points closer to the bell-margin than to the manubrium. Manu-
brium flask-shaped. Stomach pentagonal when viewed from the oral
or aboral surface of the medusa. Five simple recurved lips, with a
1 Named for Dr. \\ . II. Longley.
t
PENTAMEROUS HYDROMEDUSA FROM TORTUGAS 117
thin line of nematocysts along their edges. Entoderm of the stomach
and gonads, translucently milky. Basal bulbs of the tentacles, and
rudimentary bulbs, with a dark entodermal mass. There is an
occasional variant, as in P. pentata, with brick-red entodermal pigment
in the radial canals, tentacle bulbs, and manubrium.
The number and arrangement of the tentacles, rudimentary bulbs,
and lithocysts is very variable; even different interradial sectors of the
same medusa may differ greatly in this respect. There is often some
unevenness in the distribution of the rudimentary bulbs, but without
the extreme concentration in one sector described for Pseudoclytia
gardneri Browne.
Young Medusa. — In medusa? 2 mm. or less in diameter, the gonads
are not well-marked. In slightly larger specimens, the gonads are
distinguishable as small rounded masses on the radial canals, in about
the same position as those of the adult. There were no gonads
apparent in the smallest medusa examined, which was .8 mm. in
diameter. This medusa had eight tentacles, two tentacle bulbs, and
five lithocysts, irregularly arranged, except that there was a tentacle
at the end of each radial canal.
The polyp stages of the Pseudoclytix have not been described.
Although some search was made, no hydroids attributable to this
species were found. The presence of very early free stages, however,
indicates that the fixed stages should be found in the neighborhood
of the Tortugas.
The description of Pseudoclytia longleyi given above was prepared
from living material examined at the Tortugas. The type, and
cotypes, are deposited in the United States National Museum, and co-
types are contained in the collection of the Zoology Department of
Tulane University.
Pseudoclytia longleyi differs from P. pentata Mayer (1900) in the
following respects:
1. The presence of permanently rudimentary tentacle bulbs.
2. The larger number of lithocysts.
3. The smaller size of the medusa.
4. The situation of the gonads nearer to the bell-margin than to the
manubrium.
5. The irregularity of the arrangement of the marginal appendages.
Pseudoclytia longleyi appears to be quite close to P. gardneri Browne.
This medusa was described (Browne, 1904) from two specimens taken
in the Indian Ocean. Mayer (1910) notes that they may have been
aberrant specimens of a Phialidium. P. longleyi differs from P.
gardneri in the following respects:
8
118 M. D. BURKEXROAI)
1. The larger number of tentacles, rudimentary bulbs, and lithocysts.
2. The symmetrical arrangement of the radial canals.
3. The situation of gonads nearer to the bell-margin than to the
tnanubrium.
4. Tin- greater diffuseness in the distribution of the rudimentary bulbs.
It has been suggested that the medusa described above may be a
pentamerous variant, such as has been often noted for many lepto-
incdusae, of Phialucium carolinx Mayer. However, the following
facts seem to the writer to strongly indicate that Pseudoclytia longleyi
must be regarded as a distinct species, and not as a variant of Phialu-
cium:
1. Pseudoclytia longleyi was present at Tortugas during 1029 in
enormous numbers, while not a single specimen of the suggested
parent stock, Phialucium, was taken. The center of abundance for
Phialucium Carolina? seems to lie farther north, and it is reported by
Mayer (1910) as only occasional at the Tortugas. It is usual, so far
as the writer knows, for medusae which differ varietally from a parent
stock in the number of radial canals to appear in company with the
parent form as aberrant individuals.
2. There are strong structural differences between Pseudoclytia
longlcyi and Phialucium caroling, aside from the difference in the
number of radial canals. Owing to the difficulty of obtaining speci-
mens of P. caroling, this comparison luis been made from the descrip-
tion and figures given by Mayer (1910).
Phialucium carolinx Pseudoclytia longleyi
Bell almost a hemisphere. (1) Bell flatter than a hemisphere.
Gelatinous substance of the bell (2) Gelatinous substance of the
quite thick. bell thin and tenuous.
Rudimentary tentacle bulbs much (3) Rudimentary tentacle bulbs
smaller than the shaft-bearing often almost as large as the
bulbs. shaft-bearing bulbs.
Lithocysts with two concretions. (4) Lithocysts with one concre-
tion.
Bell diameter, 14 mm. (5) Bell diameter, 8 mm. or less.
Entoderm yellow-green. (6) Entoderm milky or red.
Marginal appendages regularly (7) Marginal appendages quite
arranged. irregularly arranged.
The writer wishes to thank the Carnegie Institution of Washington
!<>r tin- prixilege of working at the Tortugas Laboratory.
PENTAMEROUS HYDROMEDUSA FROM TORTUGAS 119
LITERATURE CITED
BROWNE, P., 1904. Fauna and Geography Maldive and Laccadive Archipelagoes.
Vol. 2, Part 3, p. 370.
GRAVE, C., AND BURKENROAD, M. D., 1929. Examination of Pelagic Organisms.
Carnegie Inst. Year Book No. 28, pp. 283-84.
MAYER, A. G., 1900. Bull. Mns. Com par. Zoo!., ll<m>ard University, 37: 53.
MAYER, A. G., 1910. Medusa of the World. Vol. 2, pp. 274-76, 278-79.
THE EFFECT OF CERTAIN ENVIRONMENTAL FACTORS
ON THE DEVELOPMENT AND HATCHING OF
THE EGGS OF BLOOD FLUKES
A. R. OXORATO AND H. \V. STUNKARD
(From the Biological Laboratory, New York University)
The digenetic trematodes have peculiarly elaborate and compli-
cated life histories. Ordinarily, sexual reproduction in a vertebrate
host produces eggs from which a free-swimming larval stage (mira-
cidium) emerges and penetrates into an invertebrate host where
repeated asexual multiplication produces a second and different free-
swimming larval form (cercaria) which either directly or indirectly,
with or without encystment, again infests the specific vertebrate host.
Although the general outlines of this complex development have been
known for many years and the life cycles of several species have been
experimentally demonstrated, the details of the particular processes
involved are as yet obscure. The factors operative in the development
and hatching of the eggs, penetration of the larvae into specific hosts,
encystment, excystment, and other vital phenomena are almost
entirely unknown.
Mattes (1926) studied the development of Fasciola hepatica, a
species which infests the liver and gall bladder of cattle and sheep,
and performed experiments to determine the factors operative in the
emergence of the miracidia from the eggs. He found that water at
a temperature of 12° C., or above, is necessary for development and
that temperatures below • 7° C. will inhibit further development,
whereas eggs can recover from exposure to temperatures not lower
than - 3° C. He found that the optimum range was between 20° C.
and 25° C., at which temperatures the miracidia required from two
to three weeks to complete their development. At a range of 14° to
18° C., they required from three to six weeks. Mattes' results indicate
also that a sudden drop in temperature may induce the emergence of
fully developed larvae. According to him. a pH of 7.5 to 8 was the
most favorable H-ion concentration for development; lowering the
pH below 6.5 caused more injury than raising it above 9; and a pH
of 5 to 7.5 was the favorable range for hatching. Mattes believed
that rain facilitates the emergence of miracidia, since it lowers both
the temperature and pH of the water. He found that the miracidia
within the eggs are positively phototrophic and that they die if the
water becomes putrid or if it entirely evaporates.
120
ENVIRONMENT AND EGGS OF BLOOD FLUKES 121
Stunkard (1923) described blood flukes from the heart and arteries
of various species of turtles and made certain observations on the
development of the parasites. Chrysemys marginata and C. picta
harbor species of Spir orchis. The flukes deposit their eggs in the
blood vessels, by means of which they are distributed throughout the
body, especially the visceral organs. The eggs rupture the capillaries
and thus gain access to the tissues, through which they work their
way to the alimentary tract or one of its derivatives, and ultimately
they are voided with the feces. When passed by the turtles the eggs
are within mucous pellets which facilitate the collection and manipu-
lation of such minute objects. The eggs of these trematodes appear
then to be particularly favorable material for study.
MATERIAL AND METHODS
Since information concerning the factors which influence early
development and emergence of trematode larvae are so important
from a purely scientific as well as a medical and economic point of
view, it seemed desirable to find out whether Mattes' results with
F. hepatica are general and whether they are applicable to the eggs of
other species. Early in October, 1929, a series of experiments was
begun to determine the relation of various factors to the development
and hatching of the eggs of blood flukes. The study was made on
eggs voided by flukes which infect the turtles named above. Since
it is not possible to make specific determination of the eggs, only
generic identification is given. These turtles voided many eggs from
October until the early part of the following January. For about a
month, from the second week of January till the second week of
February, the eggs were less numerous, but after this time they
appeared again in abundance.
The procedure throughout the study was a comparatively simple
one; the turtles were kept in separate aquaria and the eggs were
collected daily. The greenish mucous patches, voided by the turtles
and in which the eggs are embedded, were picked up in a pipette and
placed in small labelled beakers containing tap water which was
changed regularly to prevent putrefaction. By this method, a con-
stant supply of eggs was obtained and those containing larvae of differ-
ent ages and from different turtles were kept separate.
When the eggs are voided, those in the same mucous pellet vary
in their stage of development and in the color of the shell, which is
usually a darker brown in the more advanced individuals. This
variation has been explained by Stunkard (1923) by the theory that
the eggs deposited in various parts of the body of the host require
122 A. R. ONORATO AND II. W. STUXKARD
different lengths of time to reach the outside. He also noted that
development in the tissues of the turtle is very slow as compared
with later development in water, and suggested that the more rapid
later development is caused by the absorption of water and by in-
creased oxygen supply. In the present study the development of the
larva- was followed by daily observations on the eggs. It was obvi-
ou-ly necessary to wait until the miracidia became mature before
experiments could be made to determine the effects of various factors
on hatching.
I nless otherwise stated, experiments were carried on at room
temperature, since previous work indicated that it is suitable for the
development and hatching of these eggs.
The La Motte Standard colorimetric method was used to determine
the H-ion concentration of solutions.
OBSERVATIONS
The earlier observations of Stunkard on the contents of the eggs
and the course of development were confirmed and extended. A
short time after the egg is voided and before the miracidium begins
to move, vacuoles can be seen inside the shell. It is probable that
these are gas vacuoles and that the gas is carbon dioxide, a product
of metabolism of the larva. As development proceeds, the vacuoles
become larger due to fusion of the smaller ones and also to the in-
creasing accumulation of gas as the miracidium becomes more active.
When the larva is mature there is only one vacuole, often as large as
ihe miracidium itself. This vacuole was shown in Stunkard's figure
but was not described by him. The vacuole disappears as the mira-
cidium becomes increasingly active and the question naturally arises
as to what lias become of the gas and the pressure that must have
been caused by it. It seemed possible that the disappearance of the
vacuole might be associated with the opening of the shell, but in
maii\ instances the vacuole disappeared before the shell gave any
evidence of opening. \Ye are not at all certain as to the nature,
function, and fate of this vacuole.
In the mature miracidia llame cells were regularly observed; the
posterior cells arc more readily perceived than the anterior ones.
Manic cells ha\e been observed beating continuously for hours, and
counts show that they beat approximately 160 times per minute.
The miracidia while still inside the shell are sensitive to light,
which is i" !><• expected since they possess well-developed eye spots.
The light stimulus usually requires from a lew seconds to about three
minute-- bO Cause the miracidia to become active. Special precaution
ENVIRONMENT AND EGGS OF BLOOD FLUKES 123
was taken to ascertain that it was light and not heat from the lamp
that produced the activation of the larvae. The light was focused on
a thermometer placed on the stage of the microscope in the position
of the slide and after a 5-minute exposure, the mercury had risen only
0.5° C. Such a slight change in temperature, within the favorable
range, could hardly cause the rapid responses manifested by the larvae
when they are stimulated by light.
The shell bears an operculum which is invisible until it starts
to open. It is probable that this cap is preformed, since it is always
about the same size and at one end of the shell. The removal of the
operculum leaves a circular opening through which the miracidium
emerges. The escape of the larva was observed many times and the
process usually required about five or ten minutes, although occa-
sionally it may take much longer. In one instance a miracidium
required 1 hour and 15 minutes to emerge from the shell. The larvai
ordinarily come out anterior end first, but one was observed escaping
with its posterior end first. This observation may be significant,
as will be pointed out later.
When the cap opens, a fluid mass flows out of the shell along
with the larva, and this mass apparently forms an envelope of some
kind, for the miracidium has a difficult task to get out of it. Mira-
cidia have been observed for as long as four hours attempting to
escape from this viscous fluid. Many of them die after struggling
for a long time. It has been observed that the larvae move about
just as freely inside the viscous mass outside the shell as they do
inside the shell. It is possible that this substance, which gives a
protein reaction, forms a rather flexible membrane when it comes in
contact with water. Another possibility, although it is not supported
strongly by our observations, is that this is the vitelline membrane
of the egg detached from the shell and passed out of the opercular
opening with the larva. It is probable that the limiting membrane
rather than the viscosity of the fluid prevents the escape of the mira-
cidia since they appear to be continually probing into it. At first
the membrane merely gives under pressure and returns to its original
position when the pressure is released. If the procedure is repeated
often and long enough, the elasticity of the membrane is gradually
reduced until it ruptures and the larva is able to escape. It has been
noted that if miracidia are liberated into the water as soon as they
emerge from the shell they usually disintegrate, and it is probable
that during the time they are in the viscous fluid they become adjusted
to life in water. If allowed to work their way out of the viscous
mass or liberated after some time in it, they swim about so rapidly
124 A. R. OXORATO AM) II. \Y. STUXKARD
that it is almost impossible to follow them. They are active for
several hours under the conditions of observation.
In three experiments, eggs were placed in a small covered dish
containing tap water and fecal material from the turtles. In this
putrid water the embryos always died and disintegrated inside the
unopened shell. On several occasions the water on the eggs was
allowed to evaporate and invariably the shells collapsed and the
larvae died. It is clear that fresh water is necessary at all times.
Although a few miracidia emerged from their shells about three
days after their removal from the aquaria in which the turtles were
kept, the usual time required for hatching was from 5 to 7 days.
Some required longer and a few remained active in their shells for as
long as three weeks. Since the eggs were maintained under identical
environmental conditions, the variation in time of hatching must be
ascribed to other factors. Normal emergence took place at room
temperature in ordinary tap water, which was changed every few
days to prevent putrefaction. The pH of the water on the fifth,
sixth, and seventh days was always between 7.2 and 7.6. It appears
that the optimum conditions for hatching are: ordinary tap water,
room temperature, and a pH zone between 7.2 and 7.6.
Effects of Acids on Development
Acetic Acid. — To test the effect of acids on development, 74 eggs
of Spirorchis, voided during the night of November 13th, were placed
in a pH 6 solution of acetic acid the following morning. On November
16th one hatched, 43 were moving in their shells, and 30 were still in
early stages of development. On November 18th twelve had hatched,
32 were moving in their shells, the rest were still immature. On
November 20th sixteen were hatched, 30 were moving, and 28 were in
early stages of development. All of the eggs except four eventually
matured and by November 26th all had hatched except the four
mentioned above. It appears that development was not hastened or
retarded at this pH since the time required was practically the same
as when the eggs were kept in tap water. It must be recorded,
however, that all of the 70 hatched miracidia died either inside the
opened shell or immediately in front of the opening, outside of the
shell. They did not succeed in freeing themselves from the viscous
mass. The results demonstrate clearly that so long as the larvae
remain inside the unopened shell their development is not affected
by the acid, but that when the shell opens, they do not long survive.
Lactic Acid. — A similar experiment with lactic acid, pH 6, gave
practically identical results.
ENVIRONMENT AND EGGS OF BLOOD FLUKES
125
Effects of A cids upon Ha Idling
Eggs that had been kept in tap water until the miracidia were
mature and moving actively inside their shells were used for these
experiments. Since Mattes found that a slightly acid solution induced
hatching of the eggs of F. hepatica, experiments were made to determine
the effects of various concentrations of acetic, lactic, and hydrochloric
acids. The first two were selected because they are frequently present
at the bottom of ponds in nature.
Acetic Acid. — Mature eggs were placed in a solution, pH 6. Re-
sults are shown in Table I, and it is apparent that contact with acid
TABLE I
Effect of H-ion Concentration on Hatching of Eggs Containing Rapidly
Moving Miracidia *
Acetic Acid
Time of
Observance
(hours)
pHe
40 eggs
pHe.5
44 eggs
pH7*
65 eggs
PH7.3*
60 eggs
1/2..
None hatched
1 hatched but
None hatched
7 hatched
larva died al-
most immedi-
ately
1
None hatched
Larvae less
None hatched
3 more hatched
active
2
2 shells opened
No more had
None hatched
No more hatched
but larvae died
inside
emerged
3. . .
1 hatched but
No more had
3 hatched
^ more hatched
dead outside
shell
emerged
4 .
No more hatched
No more had
4 more hatched
No more hatched
emerged
6. ...
2 hatched died
No more had
3 more hatched
3 more hatched
outside shell
emerged
8
No more hatched
No more had
5 more hatched
4 more hatched
emerged
12
Larvae less active
No more had
8 more hatched
2 more hatched
emerged
24 ...
1 hatched but
1 hatched but
5 more hatched
8 more hatched
dead
died at once
30 .
No more larvae
15 hatched but
None observed
Some of miracidia
active
all were dead
were swimming
It is, of course, understood that there is little or no free acid at pH; and pH7.s.
126 A. R. OXORATO AND II. \V. STC.NKAKI)
at this concentration results in the death of the larvae. Other eggs
were placed in a solution, pH 6.5. The results (Table I). show that
this sol in i( m is also toxic to hatched miracidia. Tap water was
added to some of the above solution to bring it to a pH of 7 and
eggs containing mature larvae were placed in this fluid. The -results
are shown in Table I and indicate that at neutrality the solution is
not toxic to the miracidia, since they hatched normally. In order to
make a further test, tap water was added to some of the pH 7 solution
to raise the pH to 7.5. Kggs placed in this solution also hatched
normally.
\\ hen eggs were placed in the 6 and 6.5 solutions, the activity of
the miracidia was greatly increased. The larvae at first swam about
more rapidly and attacked the ends of the shell vigorously. After
two or three hours, however, the activity of the larva had decreased
to less than usual and in the pH 6 solution, especially, many of them
grew weaker until they died. It appears that the acid has at first a
stimulating and later a depressing and harmful effect.
Lactic Acid. — A solution of lactic acid with a pH of 6 was prepared
and tap water was added to portions of it to make solutions with a
pH of 6.5, 7, and 7.5. Mature eggs were placed in each of these
solutions and the results are shown in Table II. The effects are not
appreciably different from those obtained with similar concentrations
of acetic acid. The acid is toxic and the stronger the concentration,
the greater the toxicity.
An egg in which the flame cells of the miraciclium were beating
vigorously was placed in some of the 6.5 solution. At the end of one
hour the flame cells were beating much less rapidly and at the end of
three hours they were beating very feebly.
Hydrochloric Acid. — A series of flasks was prepared as before.
The first contained HO with a pH of 6, and by adding tap water to
port ion^ of this solution others were made up with H-ion concentrations
of pll 6.5, 6.8, 7, and 7.5. The effects of these solutions are given in
Table III. At a pH of 6, every one of the miracidia died upon coming
in contact with the solution. After 24 hours the nnhatched larva?
were all dead and after 72 hours they had disintegrated inside the
unopened shells. In the pll 0.5 solution the unhatched miracidia
were dead at the end of 48 hours. A few experiments, not recorded
in the table, were done with IK'l at pH 0.8. In one of them, 70
mature eggs were placed in the depression slide. At the end of 24
hours, .U eggs had hatched. Several of the larva- had escaped from
the viscous masses, although most of them were dead inside the
viscous material. In the pll 7 solution the eggs hatched normally
ENVIRONMENT AND EGGS OF BLOOD FLUKES
127
TABLE II
Lactic Acid
Time
(hours)
pH6
60 eggs
plU.i
62 CK«S
pH,
89 eggs
pH:.s
88 eggs
1/2...
None hatched
None hatched
N'one hatched
None hatched
1
1 hatched, but
None hatched
2 hatched
3 hatched
died
2
4 more hatched,
2 hatched
1 more hatched
1 hatched
but died
3
No more
hatched
No more hatched
2 more hatched
No more hatched
4
No more
hatched
4 more hatched
4 more hatched
No more hatched
6
1 more hatched,
6 more hatched
8 more hatched
5 more hatched
but died
8
None observed
No more hatched
6 more hatched
2 more hatched
12....
None observed
No more hatched
8 more hatched
3 more hatched
24....
22 had hatched,
but all died
without swim-
ming
6 more hatched
5 more hatched
8 more hatched
30....
No observation
No observation
At 36 hours 41 eggs
had hatched and
most of the mira-
cidia had become
free-swimming
48....
28 eggs had hatched
at this time, but
all the miracidia
had died inside the
shell or near the
opening. None
had become free-
swimming
At this time 66
eggs had
hatched and
the most of
the larvae had
become free-
swimming
and the miracidia were not killed. It appears that this acid at
concentrations of pH 6 and 6.5 is lethal to the larvae and inhibits the
hatching of the eggs.
128
A. R. ONORATO AND H. \V. STUNKARD
TABLE III
Hydrochloric Acid
Time
(hours)
pH6
76 eggs
pHe.e
76 eggs
pHi
69 eggs
PH7.5
58 eggs
1/2...
None hatched
None hatched
None hatched
In this experiment
only 1 egg hatched
in 7 days although
the larva remained
active inside the
shells
1
2 hatched, but
died
None hatched
1 hatched
2 .
3 more hatched
1 hatched, but died
3 more hatched
3
2 more hatched
None
2 more hatched
No
explanation
4
3 more hatched
2 more hatched, but
died
5 more hatched
6
5 more hatched
4 more hatched, but
died
8 more hatched
8
3 more hatched
2 more hatched, but
died
3 more hatched
12....
4 more hatched
3 more hatched, but
died
5 more hatched
24... 9
8 more hatched
rest inactive
eggs
3 more hatched, but
died
6 more hatched
48....
At this time 22 eggs
had hatched, but
larv.i1 all dead
At end of 48
hours 55 eggs
had hatched
Effects of Various Bases
A series of experiments was also carried out with the bases Na, K,
and Ca (Table IV). Solutions were prepared with concentrations of
pH 8 and 8.5. The results show that these solutions are all toxic,
their toxicity apparently increasing in the following order; KOH,
NaOH, and Ca(OH)2.
Effects of Various Temperatures
A number of experiments were made to determine the limits at
which these blood fluke larva? could exist, and also the optimum
temperature for development and hatching. If water in which mature
eggs were placed was allowed to freeze, the miracidia died although
the shells did not collapse. Fifty mature eggs were exposed to a
temperature of 0° to 2° C. for 2.5 hours. \Yhen removed the larvae
ENVIRONMENT AND EGGS OF BLOOD FLUKES
129
TAHLL- IV
Effect of H-ion Concentration and Certain Bases on Hatching of Eggs Containing
Rapidly Moving Miracidia
Time
(hours)
NaOH
pHs
66 eggs
NaOH
pHs.s
72 eggs
KOH
pH8
62 eggs
KOH
pHs.s
60 eggs
Ca(OH)2
pHs
85 eggs
Ca(OH)2
pHs.6
74 eggs
1/2...
No effect
No effect
No effect
No effect
No effect
No effect
1
No effect
No effect
No effect
No effect
No effect
No effect
2
No effect
No effect
No effect
No effect
No effect
No effect
3
No effect
No effect
No effect
3 hatched
No effect
No effect
and died,
but not as
soon as in
strong
acids
4
No effect
No effect
2 hatched,
4 more
No effect
No effect
but dead
hatched,
but dead
6
No effect
No effect
No effect
No more
2 shells
No effect
were
open, but
larvae
were
dead in
the shell
8
No effect
No effect
3 more
8 more
All quiet
No effect
hatched,
hatched
but dead
12....
No effect
No effect
2 more
2 more
All quiet
No effect
hatched,
hatched
but dead
24....
Few hatch-
Few hatch-
3 more
9 more
All dead
1 open
ed, but
ed, but
hatched,
hatched,
shell,
dead. Rest
dead. Rest
but dead.
rest dead
larva
dead in the
started to
Rest dead
dead.
shell
disinte-
Rest all
grate
dead in
the shell
and dis-
integra-
ted
were quiet, but on allowing the water to reach room temperature
again, they recovered and almost all of the eggs hatched during the
next 48 hours. Another hatch of eggs was subjected to the same
130 A. R. OXORATO AND II. \Y. STCXKARD
temperature for a period of five hours. Xone of the miracidia re-
covered from this exposure after being brought back gradually to
room temperature. Apparently, a 5-hour exposure at this temperature
range is lethal. In the next experiment 30 eggs were placed at 10° C.
for eight hours. At the end of that time the miracidia were motionless,
but they soon became active when returned to room temperature.
^cveral eggs were placed in an incubator at 25° C. for 12 hours;
during which time about forty per cent of them hatched. A large
number of mature eggs were placed at 36° C. for 24 hours. Xone of
them hatched during the time, but sixty per cent of them hatched
during the next 24 hours when kept at room temperature. Other
eggs were kept at 40° C. for 12 hours and none of the larvae recovered
from the exposure.
DISCUSSION
In the development of the eggs of Spirorchis we have described
the formation of certain vacuoles which we believe contain a gas,
probably carbon dioxide. In many respects these vacuoles resemble
the structures described by Barlow (1925) as "oily masses" in the
development of the egg of Fasciolopsis buski.
There are various theories to account for the opening of the
operculum, none of which are entirely free from difficulties and
objections. One theory postulates that the activity of the larva is
sufficient of itself to force the cap off the shell. Such an explanation
was advanced by Johnson (1920) for the hatching of eggs of Echino-
stoma revolution. It is unlikely, however, that this explanation can
apply to the hatching of Spirorchis eggs, since miracidia were observed
for several days moving actively within eggs without any apparent
effect upon the caps.
Another explanation is that the cap is cemented to the main body
of the shell and that the cement is gradually dissolved to such a degree
that the activity of the larva eventually forces the cap open. Serious
objections confront this explanation also. In the course of these
experiments mature eggs were kept for many days in water, in acids,
and in alkaline solutions without any apparent loosening of the caps.
Now if the operculum opens because the cement is dissolved by the
surrounding medium, the caps should have come off all of the eggs
eventually.
Another theory is that imbibition of water causes a high pressure
within the shell, this pressure finally forcing tin- cap off. One mira-
cidium was seen to emerge with its posterior end first, which is very
u n usual. By comparing this egg with others in the process of hatching
and with hatched miracidia which were re-entering the shell, it was
ENVIRONMENT AND EGGS OF BLOOD FLUKES 131
apparent that this miracidium actually emerged with its posterior end
first, and that it was not merely a case of a hatched larva re-entering
the vacated shell. This observation suggests that internal pressure
probably forced the cap off, and simultaneously pushed the miracidium
into the opening regardless of its position. The eggs in water tend to
become spherical, which indicates absorption of water. Although
imbibition may cause an increase of internal pressure, the factors
controlling imbibition of water are obscure.
Stunkard (1923) described two cephalic glands in the miracidium
of Spirorchis, which open near the anterior papilla. Such glands,
present in many miracidia, have been regarded by various authors
as penetration glands, the secretion of which aids the larva in its
entrance into a snail. It is possible that these glands, or others
which open slightly behind the anterior end and secrete an oily
material, becoming functional when.the miracidium is mature, produce
a substance which helps to dissolve the cement and loosen the cap.
This theory is supported by the fact that the shells do not open
until the larvae are mature, regardless of the medium in which they
are placed.
A somewrhat similar observation was made by Barlow (1925) on
the hatching of the eggs of F. buski. In that species he described the
formation of a "mucoid plug" during the development of the mira-
cidium and the application of the plug to the inside of the shell in
such a manner that it covered the operculum. According to him
this mass "seems to serve the purpose of protecting the operculum
from the action of the secretions of the miracidium." The process of
hatching was described as follows: "This is the evident action of the
miracidium as it approaches the mucoid plug. It has a substance to
deposit on the plug to erode it and allow of escape. It approaches
with caution, applies the erosive to the plug, and then contracts
vigorously several times in order to stimulate secretion. When the
plug begins to get thin at the apex of the dimple, bubbles begin to
show in a little line (Fig. 16) and then these coalesce to form a little
tube (Fig. 17). When this tubule finally opens through, the effect on
the operculum is instantaneous. No matter whether the miracidium
is in contraction or in extension at the time, the operculum flies back
on its hinge, water enters the egg, the miracidium becomes violently
excited, ciliary motion increases to more than three hundred vibrations
a minute, and the miracidium is partly extruded by the hypertonicity
of the egg contents and partly assists in its own escape." The anthro-
pomorphic interpretation of Barlow's graphic account appears to lack
support, but since conditions in the eggs of Fasciolopsis are so different
from those of Spirorchis, comparisons are not opportune.
132 A. R. OXORATO AM) II. \V. STINKARD
The results obtained from a study of the development and hatching
of the eggs of blood flukes vary considerably from those secured by
Mattes for F. hepatica. The miracidia of the latter species require
from two to six weeks to develop, whereas those of Spirorchis mature
and emerge in about a week. The factors which induced hatching
of the liver fluke eggs were not effective for those of blood flukes.
SUMMARY AND CONCLUSIONS
(1) The miracidia are positively phototrophic.
(2) The contents of the egg, which flow out of the shell with the
larva, form a membrane which prevents the miracidium from escaping
at once into the water. The protein in this substance may be the
cause of the membrane formation.
(3) Ordinary tap water is most suitable for the development and
hatching of these eggs. Putrid water is toxic to the larvae.
(4) If the water in the culture is allowed to evaporate, the larvae die.
(5) The most favorable pH zone for hatching of these eggs lies
between 7.2 and 7.6. The miracidia cannot live in a solution having
a pH lower than 6.8 or higher than 8.
(6) The optimum temperature for development and hatching is
the range 20°-25° C. Exposure to temperature of 0° C. for five hours
had a fatal effect upon the miracidia. Although they became quiescent
at a temperature of 10° C., they recovered from it on return to room
temperature. After extended exposure to temperatures higher than
40° C. the larva1 do not recover.
(7) It is postulated that conditions within the egg shell, rather
than those in the surrounding medium, cause the opening of the
operculum and emergence of the miracidium.
BIBLIOGRAPHY
BARLOW, C. H., 1925. The Life Cycle of the Human Intestinal Fluke, Fasciolopsis
buski (Lankester). Am. Jour. Hyg., Monogr. Ser. Xo. 4.
JOHNSON, J. C., 1920. The Life Cycle of Kchinostoma revolutiini (Froelich).
Univ. Calif. Publ. in Zoo/., 19: 335.
MATTES, OTTO, 1926. Zur Biologic der Larvenentwicklung von Fasciola hepatica,
besonciers iiber den Kinfluss der \Yasserstottionenkonxent rat ion auf das
Ausschliipfen der Miracidien. Zoo/. Anzeig., 69: 138.
STUNKARD, II. \Y.f 1923. Studies on Xorth American Blood Flukes. Bull. Am.
Mus. Nat. Hist., 48: 165.
Vol. LXI, No. 2 October, 1931
THE
BIOLOGICAL BULLETIN
PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY
CONTINUOUS VERSUS INTERRUPTED IRRADIATION
AXD THE RATE OF MUTATION IN DROSOPHILA
J. T. PATTERSON
Department of Zoology, The University of Texas, Austin, Texas
It has been shown by Hanson and Heys (1928) for radium and by
Oliver (1930) for X-rays that the rate of mutation in Drosophila, as
measured by the percentage of sex-linked lethals produced, is determined
by the strength of the dose used in the treatments. In general their
results justify the conclusion that the rate is directly proportional to the
dosage employed. This conclusion is based on results obtained in
experiments in which the treatment was given continuously, and in
which the only variable factor involved was a difference in the length
of exposure. If all other factors are kept constant, the length of the
treatment will determine the amount of ionizing radiation to which the
flies are exposed. Under these conditions of experimentation they find,
for example, that if the time of exposure is doubled, the number of sex-
linked lethals produced will also be doubled.
It would seem to be a matter of interest to determine what effect
on the mutation rate would follow if the dosage were given in fractions
at regularly spaced intervals, instead of continuously as a single dose.
Would the summation of the several spaced treatments give a cumulative
effect on the rate of mutation, or would the rate be the same as for a
single treatment in which the same total dosage was used ?
The writer has carried out a series of experiments with a view to
finding an answer to this question. In all, 16,963 F., cultures have been
developed from the several different experiments performed. The
method employed was the same as that used by Oliver (1930) and
others. The treated flies were tested by Muller's C1B method, which
has been fully described by several different writers. The chief point to
emphasize is the one already mentioned by Oliver: not all F2 female
cultures are due to point mutation lethals ; some result from chromosome
abnormalities. Genetic tests were made for the detection of such cases,
and those that were found have been excluded from the data listed in
133
10
134
J. T. PATTERSON
the accompanying table. Four separate sets of experiments were car-
ried out, and all of the treated and control Pt males for a given set were
taken from a single culture bottle. The treated flies belonged to a wild-
type strain of Drosophila melanogaster that had been inbred for many
generations, over a period of two years.
TABLE
Groups
of Pi
Males
Tync
of
Filter
Tar-
get
Dis-
tance
Nature
of
Treatment
Time
Interval
Total
Time
Total
Dose
in r
Units
Numb-
er of
Fi
Tubes
Lethal
Muta-
tions
Percent-
age of
Lethals
• nt.
1
alum.
12
con-
1 0 min.
1654
971
4')
4.95
tinuous
2. . . .
alum.
12
spaced
12 hr.
16 min.
1654
993
62
6.15
3 ....
alum.
12
spaced
(. hr.
16 min.
1654
981
71
7.14
4
card
12
con-
8 min.
2558
518
39
7.41
tinuous
5
card
12
spaced
12 hr.
8 min.
2558
345
45
12.95
6.. . .
controls
986
1
0.10
7 ....
card
23
con-
,
10 min.
1234
863
28
3.11
tinuous
8
card
23
spaced
24 hr.
10 min.i
1220
876
31
3.33
9.. ..
card
23
spaced
12 hr.
10 min.i
1221
936
40
4.06
10....
card
23
spaced
8 hr.
10 min.i
1219
856
34
3.76
11....
card
23
spaced
1 hr.
10 min. ±
1220
1014
32
2.94
12.. ..
card
23
spaced
30 nun.
10 min.±
1234
962
33
3.22
13....
alum.
12
con-
S min.
864
919
22
2.19
tinuous
14
alum.
12
spaced
1 min.
S n i i n . ±
872
980
21
1.93
15
controls
951
2
0.21
16
alum.
3
con-
12 hr.
radi-
544
58
10.44
tinuous
um
17.. ..
alum.
3
spaced
12 hr.
12 hr.
radi-
452
48
10.40
um
18.. ..
cuiil n>K
453
1
0.22
The first series of experiments, which was of a preliminary char-
acter, gave results that were not decisive, both because the numbers
employed were too small, and because the controls proved to be in-
adequate, hi these tests two groups of larvae were X-rayed, one con-
tinuously and the other intermittently, and two groups of adult males
were treated in the same manner. The intermittent exposures were
given at regularly spaced intervals, and their sum in each instance was
exactly equal to the time of the continuous exposure. The results
showed that the groups given the spaced exposures yielded in the F..
generations a higher percentage of sex-linked lethals than was found in
the descendants of flics that had been given continuous treatments. The
difference, however, was scarcely statistically significant. It was there-
tore.' decided to repeat the experiments on a larger scale, using longer
treatments and varying the intervals in the spaced exposures.
RATE OF MUTATION IN DROSOPHILA 135
In the second set of experiments six different groups of males (all
from the same culture bottle) were used. Five of these were treated,
and one was used as a control (Table, group 6). The first five groups
were treated as follows: Group 1 was exposed for sixteen minutes with
the machine operated at 50 kv., peak 10 ma., target distance 12 cm., and
a 1-mm. aluminum filter. The total dosage in " r " units (column 7)
was calculated from sample readings on a Yictoreen dosimeter, taken
iust after the treatment had been completed. Group 2 was given six-
teen exposures, each of exactly one-minute duration, at regular intervals
of twelve hours. Group 3 was given thirty-two exposures, each of
thirty seconds' duration, at regular intervals of six hours. Group 4
was treated continuously for eight minutes, with the machine operated
at the same level as before, but with a card filter of .28 mm. thickness
substituted for the aluminum filter. Finally, group 5 was given sixteen
exposures, at regular intervals of twelve hours, and each of thirty
seconds' duration.
The percentages (corrected for controls) of the tested sex-linked
lethal mutations are shown in the last column of the table. It is obvious
that groups 2 and 3 should be compared with group 1 and that group 5
should be compared with group 4. Calculations show that the dif-
ference between group 1 and group 2 is .012 ± .007. The difference is
less than twice its own probable error, and is therefore not significant.
If we compare group 3 with group 1, the difference is found to be
.0219 ± .0073. Here the difference is exactly three times its own prob-
able error and may or may not be significant. If groups 4 and 5 be
compared in a similar manner, the difference is found to be .0554 ±\
.0138, or slightly more than four times the probable error. It would
therefore seem to be statistically significant.
The differences noted above may have been due to chance variations,
or they may have resulted from experimental error, incident to starting
and stopping the machine. If the sum of the several spaced exposures
does not exactly equal (in the total amount of radiation) the time of the
continuous exposure, differences in the mutation rate would necessarily
occur. In giving the treatments every care was taken to make the two
equal. The same voltage, milli-amperage, and target distance were used,
and the intervals of time were carefully determined by means of a stop-
watch. The total dosages (groups 1 to 5) were calculated, as stated
above, from sample readings taken on the dosimeter just after the flies
had been exposed, with the machine operated at the same level as was
used in giving the treatments. But this method might be a source of
error. It was therefore decided to arrange the apparatus so that the
dosimeter readings could be taken at the time of making the exposures.
136 J. T. PATTERSON
Numerous trial tests showed that so long as the machine was operated
at the same level, one obtained very consistent readings.
The next series of experiments (groups 7 to 15) were conducted
under the conditions just mentioned. The table shows the various de-
tails of the several experiments, except that the length of the individual
exposures and the exact manner of calculating the total amounts of
radiation as measured in r units is not revealed. \Yith reference to the
latter point, it may be stated that it was found necessary to vary, slightly
the length of the last exposure in the series of any given spaced test.
This was done in order to make the total number of r units correspond
as nearly as possible to that used in the continuous exposure. As an
example, in group 7. the flies were exposed for ten minutes, and dosi-
meter readings were taken every thirty seconds. Calculations based on
these twenty readings gave a total dosage of 1234 r units of radiation.
The flies of group 8 were given eight exposures, each of one and a
quarter minutes' duration, at regular intervals of twenty-four hours.
Two readings were taken during each exposure, and after the first seven
treatments had been completed the total dosage up to that point was
determined. From the figure obtained, it was found necessary to
shorten the last exposure five seconds, in order to treat this group of
flies with a total dosage1 approximating that given to group 7. This gave
a total of 1220 r units. The succeeding groups were handled in a similar
manner. The deviation from the total time set by the continuous treat-
ment was plus or minus about five seconds in each spaced test.
( Iroupx S to 12 form a series in which the time interval for the spaced
exposures was gradually shortened from twenty-four hours to thirty
minutes. The corrected percentages of sex-linked lethal mutations run
as follows: continuous (7), 3.11; twenty-four-hour interval (8), 3.33;
twelve-hour interval (9), 4.06; eight-hour interval (10), 3.76; one-hour
interval (11), 2.94; thirty-minute interval (12), 3.22. From this it will
be seen that the percentages vary from 2.94 in group 1 1 to 4.06 in group
9. If one compares the percentage of lethals for the spaced exposures
(groups 8 to 12) with that for the continuous exposure (group 7). one
finds that the differences are not significant. The percentage of group
9 shows the widest divergency from that obtained in group 7 (3.11 ), but
calculations show that the difference is .0105 ± .0060. which is less than
three times the probable error, and is therefore not statistically signifi-
cant. Furthermore, the percentage of lethals obtained in the spaced
exposures of group 1 1 is actually lower than that for group 7. Finally,
in groups 13 and 14. in which there were eight-minute treatments, the
spaced treatment again gave a lower percentage than was produced by
the continuous exposure.
RATE OF MUTATION IN DROSOPHILA 137
From the results obtained in this set of experiments, one may con-
clude that to give the treatment in fractional doses does not affect the
rate of mutation in any way different from that found after a continuous
exposure of equal strength had been used. The differences noted are
due either to chance variation inherent in the material, or to small un-
controllable fluctuations in the voltage and milli-amperage of the ma-
chine. The ideal source of radiation for such a test would have been
radium, because the amount of variation in dosage when this substance
is used is negligible. Up to the time at which the above experiments
were completed (July, 1930) radium was not available, but early last
fall, through the kindness of Dr. W. C. Curtis of the National Research
Council, 123 mg. of radium were loaned to the laboratory, and this was
used in the following test (groups 16 and 17).
The arrangements for giving the radium treatments were such that
it was possible to avoid any appreciable error in the time factor. Two
groups of Pj male flies were exposed ; group 16 was given a continuous
treatment lasting twelve hours, while group 17 was treated with twelve
exposures each of exactly one hour, at regular intervals of twelve hours.
The F.2 cultures gave for continuous and spaced treatments 10.44 per
cent and 10.40 per cent of lethals, respectively. The difference is ob-
viously insignificant.
CONCLUSION
The conclusion that one may draw from these experiments is that
the rate of mutation, as measured by the number of sex-linked lethals
produced, is the same whether the treatment is given in one dose or in
several fractional doses. This is true so long as the total amount of
radiation under each of the two conditions is made equal.
The effects of continuous and spaced radiation have been studied
by several different investigators, but the only reference found, in which
their effects on the rate of mutation are mentioned, is a paper by
Serebrovsky and Dubinin (June, 1930), dealing with the production
of mutations in Drosopliila by X-rays. They report (p. 260) that
continuous and interrupted exposures were given, and in this connection
state that, " It is clear that lengthening the time of exposure with inter-
ruption or without raises the percent of mutation." One can not tell
from this brief statement whether the percentages were the same or
different under the two methods of treatment.
Most of the other papers concerned with methods of administering
irradiation doses deal with embryological or histological materials, and
they are therefore of no special interest in this connection. One of the
latest papers along these lines is by F. G. Spear (May, 1931). Spear
138 J. T. PATTERSON
used radium and studied and compared the delayed lethal effect on tissue
cultures in vitro, after spaced and continuous irradiations had been given.
He used the subcultivation method and concludes from his results that
it is immaterial whether the irradiation is given in one or in several
fractional doses — the delayed lethal effect is the same.
Austin, Texas,
June 1, 1931.
REFERENCES
HANSON, F. B., AND FLOKE.NCE M. HKVS, 1928. The Effect of Radium in Pro-
ducing Lethal Mutations in Drosophila melanogaster. Science, 68: 115.
OLIVFR, C. P., 1930. The Effect of Varying the Duration of X-ray Treatment
upon the Frequency of Mutation. Science, 71: 44.
SERERROVSKY, A. S., AND N. P. DIHIMX, 1930. X-ray Experiments with
Drosophila. Jour. Hcrcd., 21: 259.
SPEAK, F. G. 1931. The Delayed Lethal Effect of Radium on Tissue Cultures in
vitro — Comparison of Continuous and Spaced Radiation. Proc. Roy. Sac.,
Ser. B, 108: 190.
GENETIC EVIDENCE FOR DIPLOIDISM OF P,I PARENTAL
MALES IN HABROBRACON
MAGNHILD M. TORVIK
(From the Department of Zoology, Unircrsity of Pittsburgh)
I. INTRODUCTION
According to the theory of Dzierzon (1845), drones of the honey-
bee arise from unfertilized eggs. Within recent years this theory has
been extended and developed into the general conception that males in
Hymenoptera are haploid. It is unnecessary, here, to review the ex-
tensive literature dealing with this subject. Despite the evidence for
parthenogenetic production of drones, many bee-breeders still insist that
certain males in hybrid broods show paternal traits. If Dzierzon's
theory is correct, how can this be possible?
The subject of sex determination in Hymenoptera assumes consid-
erable interest to the geneticist from another point of view — that of the
theory of genie balance as emphasized by Bridges (1925).
Investigations on the parasitic wasp Habrobracon juglandis (Ash-
mead) have served to clear our conceptions to some extent although
much still remains to be done. Significant contributions already pub-
lished may be briefly mentioned.
It was first shown in 1921 (Whiting, P. W., 1921) that certain males
(patroclinous) in Habrobracon, contrary to expectation, occasionally
show paternal traits and, therefore, do not arise from unfertilized eggs.
Subsequently, Anna R. Whiting found (1925) that such males were
often abnormal and almost completely sterile.
The occurrence of new mutations made it possible to show (Whiting,
Anna R., 1927) that these irregular males (biparental) resemble their
sisters in that they inherit dominant traits from both parents.
It was at first supposed that they might be haploid mosaics. This
idea was precluded (Whiting, Anna R., 1928) by the fact that, when the
dominant members of two allelomorphic pairs affecting one and the
same part of the body (the wings) were contributed by opposite parents,
both dominant traits appeared. This indicates, though it does not prove,
diploidism.
Still better evidence of diploidism of biparental males appeared
(Whiting, Anna R., 1928) when a factor pair was used in which
139
140 MAGXHILD M. TORVIK
dominance was reversed according to the structures affected. In the
series of genes allelomorphic to orange (eye color), "ivory" has eyes
and ocelli white while in " light " eyes are black and the ocelli, although
of reduced pigmentation as compared with the normal black of wild
type. are. nevertheless, dark brown in both males and females of pure
" light ' stock. Light-ivory compound females have eyes black and
ocelli white, thereby exhibiting dominance of "light " in the former, of
" ivory " in the latter. Their biparental brothers resemble them in this
respect and thus demonstrate a duplex condition of this factor. Such
a combination would be impossible in a haploid male.
Another indication of abnormal chromosomal constitution of bi-
parental males is the fact that, occasionally, daughters that are almost
completely sterile and often morphologically abnormal are produced
(Whiting. Anna R.. 1925).
The question as to why biparental males, if diploid, are. nevertheless,
males still presents itself. It has been suggested that there may be a
sex chromosome for which these males are simplex. The experiments
summarized in the present paper were primarily planned for the purpose
of locating such a chromosome by genetic test and, in addition, for
studying the composition of daughters of biparental males.
The writer is indebted to Professor P. \Y. Whiting, at whose sug-
gestion the investigation was begun, for guidance as the work progressed.
Special thanks are due the co-workers on Habrobracon for supplying
mutant stocks. Acknowledgment should be made in particular to the
Committee on Effects of Radiation on Living Organisms (National
Research Council) who. by a grant to Professor Whiting, have fur-
nished technical help which has relieved the writer of much time-con-
suming labor and thus made possible a greater output of results.
II. MATKRIAL AND METHODS
Since the purpose was to locate a sex-linked factor, new mutations
were tested with this in mind. In most cases eye color (orange locus)
was brought into the cross in such a way, with respect to the mutation,
as to serve as an indicator of biparentalism ; a male with a dominant eye
color being crossed to a female with recessive eye color and the Yl males
with dominant eye color being taken as biparentals. It was not always
possible to do this since no light eye colors are present in some of the
stocks and, in cases thus far tested, related males and females must be
crossed in order to obtain biparental males (Whiting, Anna R., 1925).
In these stocks the dominant and recessive allelomorphs of the mutated
locus had to be used alone.
The crosses may be divided into five main types involving: 1. nor-
DIPLOIDISM IN HABROBRACON 141
mally inherited, recessive mutations in Lancasterized stocks ; 2. recessive
mutations in No. 11 and Minnesota stocks; 3. female sterile mutations;
4. interacting mutations; and 5. incompletely dominant or incompletely
recessive mutations.
The triploid nature of the daughters of hiparental males was tested
by using some of these same mutations in a manner which will be de-
scribed later.
III. PRESENTATION OF DATA
.-/. Production of Biparental Males
1. Crosses Involving Normally Inherited, Recessive Mutations in Lan-
casterized Stocks.
Lancasterized stocks have all been graded up to a stock derived from
a wild female taken in Lancaster, Pennsylvania in 1919. The method
of inheritance by biparental males of certain normally inherited recessive
mutations in these stocks has been previously discussed (Whiting,
Anna R., 1927). It was shown that, when a female homozygous for
the recessive mutant gene was crossed to a male carrying the normal
allelomorph, there were produced not only recessive haploid sons but also
a few sons with the dominant character. If females, homozygous for
the normal allelomorph but marked by some other recessive factor (eye
color, usually) were crossed to males, carrying the recessive mutant gene
and the dominant marking factor, the biparental males showed only the
dominant characters.
A series of quadruple allelomorphs affecting eye color (0, oj , o and
o>) and three pairs of allelomorphs affecting the wings (IV , w), (R, r)
and (D, d) were shown to be inherited in the above manner (Whiting,
AnnaR., 1927).
The fact that of five black-eyed biparental males, obtained from
crosses of orange wrinkled (ooww) females by type (OW} males, one
had wrinkled wings, seemed to indicate something unusual about the in-
heritance of wrinkled wings. As shown in Table I this cross was re-
peated and 20 biparental males were obtained, none of which had
wrinkled wings. The wrinkled wings of the above-mentioned male
were no doubt, as they so often are, the result of accident of growth.
As shown further in Table I, three other loci were tested by re-
ciprocal crosses. Long, /, affects both the wings and antennae ; eyeless,
cl, and small eye, ks, or extremely small, ke, (the latter 3 stocks kindly
sent by Dr. Wilhelmina F. Dunning) affect the appearance of the eyes
and shape of the head. All three factors were inherited by biparental
males in the regular manner described above.
142
MAGNHILD M. TORVIK
TABLE I
Crosses Involving Normally Inherited, Recessive Mutations in Lancasterized Stocks
Parents
I'rnneny
Sect inn
9
<?
MatinKs
Bipa rental
d"d"
Impaternate
0"^
9 9
1 a
ooww
+ (Xo. 1)
84
19
1219
2220
ooww
Y
11
1
187
322
OOU'W
+ (Xo. 31)
23
0
248
476
ooww
/
11
0
109
208
ooddww
+
3
5
43
93*
b
i'n
w
50
73
2009
1313*
1 a
o«'o-7(L)/
Y
7
?
87
9'
o'o' 11
Y
9
10
42
58
0*0*11
+ (Xo. 31)
2
0
9
11
1,
o'o' (No. 17)
/
6
2
68
28
oioi
/
20
8
436
479
cc
/
28
18
166
1()1
3 a
eld
+
20
1
2 OS
743
elel
wa
1
6
11
32
b
oo (Xo. 3) .
el
6
5
164
178
4 a
o'o'k'k*
+ (Xo. 1)
21
41
122
219
o'o'k'k'
o (Xo. 3)
3
3
28
31
b
k'k*
k>
3
3
12
11
Total
308
197
5 1 53
(.705
' Taken from Table I— Whiting, Anna K., 1927.
2. C.'rosscs Involving Recessive Mutations in No. 11 and Minnesota
Stocks.
An attempt was made to obtain biparmtal males by crussing tapering,
/</, (antenna') males of Minnesota Yelli>\v stock to orange-eyed females
of Xo. 3 Lancaster stock, but none were obtained. Attempts have also
been made to cross No. 1 1 individuals, Iowa City stock, to wasps from
Lancaster stocks with the same results. Therefore, in dealing with
mutant genes in these stocks, it was necessary to make the crosses be-
tween individuals of the same stock and to omit the eye color indicator.
It is interesting to note that biparental males do occur when crosses are
made within these stocks but do not with outcrossing. a result in agree-
ment with previous findings for other stocks (Whiting, Anna \\.. 1925).
D1PLOIDISM IN HABROBRAC ON
143
Table II gives a summary of the data obtained. Inheritance of
three mutations: tapering, ta, (antennae — Minnesota Yellow stock);
wavy, wo, (wings — No. 11 stock) and semilong, si, (antennae and wings
TABLE II
Crosses Involving Recessive Mutations in No. 11 and Minnesota Stocks
Parents
i
PniRonv
Section
9
<?
Matines
Bi parental
cTcf
ItupaUTiiute
cTcf
9 9
1
oo (No. 3)
ta (Minn )
10
o
7 60
431
/a/a (Minn.)
My (Minn.)
23
30
608
487
2
wawa (No. 11). ...
wawa (No. 1 1) ....
+ (No. 11)
+ (No. 11)
36
2
1
383
18
2
elel (?)
wa (No. 11)
1
6
11
3^
3
si si (No. 11)
+ (No. 11)
17
64
205
184
Total
89
101
1485
1136
— No. 11 stock) was tested. The data show that wavy females usually
produced only unfertilized eggs, though in many cases matings were
observed. However, one biparental male was obtained from a wavy
female. All three of these mutant genes were found to be inherited
in the usual way by biparental males.
3. Crosses Involving Female Sterile Mutations.
Several of the mutant genes in Habrobracon seem to have a rather
deleterious effect when present in the homozygous condition in the
female. In many eases the female has to be given stung caterpillars
upon which she is then able to feed and to oviposit, but such females are
never as viable as type stock females.
When certain of the genes are present the deleterious effect becomes
so great that the females are completely, or almost completely, sterile
although the gonapophyses are of normal external appearance.
In dealing with these factors it was necessary to use heterozygous
females and to assume that on the basis of chance some of the biparental
males would carry the mutated gene. Haploid males with the mutant
genes are fertile but usually have more or less difficulty in mating and,
therefore, were rather difficult to test.
144
MAGXHILD M. TORVIK
Table III gives a summary of data obtained from crosses involving
four such factors. Fused, /, causes fusion of antennal and tarsal seg-
ments and an indentation near the apexes of the wings. Miniature, in.
affects the whole wasps but especially wings and antennae. Beaded, b,
TABLE III
Crosses Involving Female Sterile Mutations
Section
Parents
Matings
Progeny
9
c?
Bi parental
d"d"
Impaternate *
6" rf1
9 9
Dominant
Recessive
1 a
1)
oo* Ff. .
+ (Xo. 1)
/
33
15
33
19
308
276
269
111
451
210f
oo* Ff. .
o*o*rr
f
f
4
1
8
2
124
1
63
6
oo (Xo. 3) ...
2 a
b
0*0* Mm. . .
I
•> (Xo. 3)
21
7
8
0
279
25
173
7
503
144
OoiMm. . . .
o*o*rr
m
m
om
12
1
3
0
2
3
99
14
63
219
33
32
00
0*0*
3 a
b
oio*Bb
+ (Xo. 1)
12
6
63
20
181
00
b
b
14
0
2
221
229
oo (Xo. 3) ...
4 b
del
sp
sp
4
10
0
5
24
128
132
247
oo (X'o. 3) ...
Total
is;
88
1625
746
2450
* Classified as dominant or recessive according to factors/, m, b, sf> for Sections
1, 2, 3, 4 respectively.
t 114 of these were fused, ff.
causes swelling of the leg segments. Spread, .v/1, causes the wings to be
held down and spread out at the sides and produces a light spot on each
side of the thorax.
Biparental males inherited these mutant genes in the usual way.
4. Crosses Involving Interacting Mutations.
Mutations at the three loci (orange, cantaloup and maroon) af-
fecting eye color are, as one would expect, complementary to one an-
other in the effect which thev have uixni eve color.
DIPLOIDISM IN HABROBRACON
145
Crosses were made (Table IV) in such ways that the dominant of
one locus was brought in with the recessive of another, by one parent
while the allelomorphs were brought in by the other parent. For ex-
ample, ivory females (o'o'CC) were crossed with cantaloup (Of) males.
In addition to ivory-eyed males and black-eyed females, 25 black-eyed
males were obtained. These biparental males, like their sisters, re-
ceived the dominant allelomorph to cantaloup from their mother, the
TABLE IV
Crosses Involving Interacting Mutations
Parents
Progeny
Section
9
d"
Matings
Biparental
tfc?
Impaternate
efcf
9 9
1 a
cc . .
o
53
1
273
698
cc
+
28
18
166
191
b
ol'o1'
c
10
25
106
73
00
c
35
17
305
399
2 a
mama
o
9
11
52
140
mama
+ (No. 1)
8
7
35
53
b
oo (No 3)
mu-
20
36
431
340
Total
163
110
1368
1894
dominant allelomorph to ivory from their father. These comple-
mentary dominant factors produced black eye color.
Reciprocal crosses involving the two loci orange and cantaloup and
other crosses involving the two loci orange and maroon, were made.
In all cases biparental males showed, by having type (black) eyes, that
both dominant factors were present in their chromosomes.
5. Crosses Involving Incompletely Dominant or Incompletely Recessive
Mutations.
An example of a factor pair with dominance reversed according to
structure affected was cited in the introduction. Biparental males were
found (Whiting, Anna R., 1928) to resemble their sisters in showing
dominance of " ivory " in ocelli but of " light " in eyes, therein- demon-
strating the presence of the two allelomorphs ;< light ' (ocelli) and
:' ivory " (eyes) at the orange locus.
Inheritance of four other genes: dahlia, o'1, (eyes); stumpy, st,
146
MAGXHILD M. TORVIK
(legs); yellow. 1'. (antennae) and short, sli. (wings) which produce
distinct heterozygous types, has been studied (Table V).
Dahlia, o'1, is one of the quintuple allelomorphs (O, ol , o'1, o, o')
of the orange locus. It was noted that dahlia-orange (o''-o) and, more
TAHLE V
Crosses Involving Incompletely Dominant or Incompletely Recessive Mutations
Parents
Progeny
Section
9
<f
Matings
Biparental
cPd1
Impatemate
cfef
9 9
1
oo (Xo. 3)
o'fl'
,'
od
17
37
10
69
112
1080
118
741
'
o'olslst
o ' \o. 3)
10
2
27
45
o'o'slst
+ (Xo. 1)
7
0
21
39
3 a
o'o'YY
0'0''(«: IT
o ' No. 3)
+ (Xo. 1)
10
41
0
93
38
5<><)
89
833
1)
oo (Xo. 3)
Y
13
1
190
319
o'o' (No. 1 7 )
V
84
70
801
95 i
4 a
onslish
+ (XTo. 1)
171
81
SI 27
3572
ooshsh
+ (Xo. 11)
4
0
74
42
b (1)
«'(>' (Xo. 17)
osli
26
7
360
321
'
fl'O' (Xo. 1 7"!
osh
32
0
332
271
(3)
<>'{>> (Xo 17)
osh
40
52
730
5s 7
Totals
4') 2
385
7491
7<>27
especially, dahlia-ivory (o''-o') compound females were much lighter
than homozygous dahlia females.
Crosses such that dahlia-ivory (o''-o'} and dahlia-orange (0rf-o)
compound biparental malo would be obtained, if possible, were made
(Table Y). Seventy-nine males were found which closely resembled
their sisters, the compound (o''-o and o''-o' ) females, and were strikingly
different from ivory or dahlia or orange haploid males. It seems fair
to assume, since they also bred like biparental males, that these males
must have been heterozygous and, therefore, duplex for eye color.
Eye color was always used as an indicator of male biparentalism in
dealing with the other genes which affect variable traits.
The mutant gene stumpy, st, (legs) is an incomplete recessive.
Stumpy wasps have extreme crowding and irregularity of tarsal seg-
ments so that, unless close examination is made, the legs appear to end
DIPLOIDISM IN HABROBRACON
147
with the tibia*. Heterozygous females show tarsal segments of ap-
proximately normal length hut irregularly set together. The majority
of these females, moreover, possess a new structure (an added spur on
the prothoracic metatarsi resemhling the prothoracic tihial spur) never
found on type wasps.
Two biparental males were ohtained (Table V) from crosses of
stumpy females by type males. Both these males had the metatarsal
spurs on both prothoracic legs and must, therefore, have been hetero-
zygous for stumpy.
The factor yellow, Y, (antennae) when wasps are reared under
standard conditions (30° C.), changes the color of the three basal seg-
ments of the anteniue from the normal black to a clear yellow. The
character is dependent for its typical expression upon this constant high
temperature.
Reciprocal crosses were made (Table V). Wasps of those counts,
only, in which biparental males appeared were preserved in alcohol as
it was considered that these were reared under the most similar condi-
tions. Later the preserved wasps were graded according to the scheme
(devised by Lysbeth Hamilton Benkert) shown in Table VI.
TABLE VI
Frequency Distribution of ]\'asf>s of Various Genetic Compositions According to Grade
of Yellois) (Antennae)
Class Values*
Experiment
No.
Genetic Composition
Mean
Value
1
1.5
2
2.5
3
4
5
I
00' Fv 9
42
36
59
16
18
2.6
GVFjc?
17
15
17
2.0
olytf
120
40
1(?)
1.2
II
Oo{(o) Yy 9
64
85
21
2.4
00*'(0)Fycf
4
23
50
5
1.9
oFtf
9
24
2.7
0Fd"
32
62
2.7
Scheme of grading devised by Lysbeth Hamilton Benkert.
* 1 — joints all dark.
2 — 3 joints yellow, slightly sooty.
3 — 3 joints clear yellow.
4 — 3 1 joints clear yellow.
5-4 joints clear yellow.
The data show clearly that the biparental males are intermediate
between the two types of haploid males (type and mutant) with respect
to this character. The biparental males also differ somewhat from their
148 MAGNHILD M. TORVIK
heterozygous sisters. Since females are normally lighter than males
of the same stock this difference between heterozygous females and
heterozygous males may be explained as a secondary sexual trait.
Variation in the character short, sh, (wings) depends upon multiple
factors but there is, evidently, one main genie difference distinguishing
short stock from type. Short overlaps with normal, especially at lower
temperatures. Higher temperatures during development increase the
difference from type.
In Experiment I orange short (No. 28) females were crossed to
type (No. 1) males and black-eyed males were selected as biparentals.
To make it possible to recognize biparental males on the basis of eye
color in the reciprocal cross, it was necessary to use an ivory (recessive
to orange) stock (No. 17) instead of the black (dominant to orange)
stock ( No. 1 ) .
Since short is such a variable character, it was considered necessary
to make measurements of the wings and to treat the data statistically.
\\hen biparental males were obtained, these and equal numbers of
brothers and sisters from the same count were preserved in 95 per cent
alcohol. Later the right mesothoracic wing and the head of each wasp
were removed and placed on a slide. The parts were covered with
separate pieces of cover glass and kept moistened with alcohol. Camera
lucida measurements were made (magnification 50 X) of the costal
margin of the wing from the tip of the tegula to the end of the radius
(point of fusion of K^ and /x",). This may be called costal length,
c.l. Similar measurements were taken of head width at the widest point
including the eyes, h.ic. The mean ratio of costal length to head width,
c.l./h.w., was used as a 1 KIM'S for comparing the various types. Table
Y1I presents a list of the calculated means and mean differences.
Data given under Experiment I (Table VII ) have been previously
discussed (Torvik, Magnhild M., 1929). The black males, though they
came from mothers homozygous for short, showed a significantly
(4.23 X S.E.) greater mean c.l./li.ic. ratio than that of their short
brothers, demonstrating the presence of the type gene (Sh} in their
chromosomes, "Wings of females average somewhat larger than wings
of males of the same stock. The mean c.l. h.w. ratio of the hetero-
zygous females (Sh .*>/;) was greater (6.34 X S.E.) than that of their
biparental brothers, showing that these males must also be carrying
short (sh). They must, therefore, be duplex for this factor (SIi sJi).
Lower mean f././//.?tp. ratios were obtained for both biparental males
and heterozygous females of Experiment II than for those of Experi-
ment I. This difference indicates that short is a multiple factor char-
acter and that stocks No. 1 and No. 17 differ with respect to some of the
factors.
DIPLOIDISM IN HABKOBRACON
149
In Experiment II \vas]>s were first reared at the temperature of the
former experiment (26°-27° C.) (Table V, line 1). Later an attempt
was made to start the voting at this temperature and then to shift the
larvae to another incubator running at a higher temperature (35° C).
As seen in Table V, line (2). from 32 crosses no biparental males were
obtained and many dead larva; were observed in vials thus transferred.
The cross was repeated at Woods Hole during the summer of 1929
and wasps which were later measured were obtained (Table Y, line 3).
The incubator at Woods Hole was kept at 30° C. rather than 26° -27° C.
Also, these wasps were reared during the summer, when taking the vials
from the incubator to transfer the mothers would have a less cooling
effect upon the larvae than the same treatment of larva? reared during
the winter. These facts help to explain the differences in mean ratios of
wasps of Experiment I and Experiment II.
TABLE VII
Mean Ratios, Costal Length to Head Width, for Wasps of Various Genetic Compositions
Exp.
No.
Genetic Composition
Group
-f±±S.E.
li.u-.
Group Differences
I
OoShsh 9
a
3.551 ± 0.025
a-b
0.215 ±0.030
OoShsh c?
b
3.336 ± 0.023
b-c
0.121 ±0.028
osh d1
c
3.215 ±0.017
II
ooshsh 9
d
3.344 ± 0.01Q
d-g
0.158 ± 0.026
oWShSh 9
e
3.590 ± 0.020
e-g
0.404 ± 0.028
oo'Shsh 9
f
3.439 ± 0.014
f-g
0.253 ± 0.023
ooiShsh<^
g
3.186 ± 0.018
h-g
0.259 ± 0.023
o'.SVfc?
h
3.445 ± 0.014
f-d
0.095 ± 0.024
e-f
0.151 ± 0.024
III*
OSh (No. 1)
3.566 ± 0.018
/'-/
0.006 ± 0.019
osh
3.182 ± 0.01S
e-d
0.246 ± 0.026
e-h
0.145 ± 0.024
c-g
0.029 ± 0.025
h-b
0.109 ± 0.027
* Data collected by Dorothy A. Binns.
In addition to the measurements of the three sorts of offspring—
ivory males, orange males and orange females — of Experiment II.
homozygous orange short (No. 28) females and ivory (No. 17) females
were also measured. An interesting fact was brought out by these
measurements (Table VII), namely, that the c.l./h.w. ratio is always
greater for the female than for the same type of male. Heads of stock-
No. 17 males and females were approximately the same average width
(male head 0.014 cm. narrower by camera lucida measure at X 50), but
the male wing was shorter than the female wing (0.352 cm.). For this
11
150 MAGXHILD M. TORVIK
reason the male ratio of c.l.'h.u'. would be proportionately lower than
that of the female.
This point must he borne in mind in interpreting the data. It ac-
counts for the fact that there is no significant difference between the
mean c.l./h.iv. ratio of heterozygous females and of No. 17 (type wings)
males while the mean c.l./h.zv. ratio of the heterozygous, biparental
males is very much lower than either. The latter ratio is lower than the
mean cJ./h.iv. ratio of the heterozygous females, presumably of the same
composition, because comparison is being made between males and
females. It is lower than that of the No. 17 males because they are
carrying only the gene for normal wings, while the biparental males
show in this way that they are also carrying the gene for short wings.
The data also indicate (Table VII) that short is more dominant
than recessive. The heterozygous females are more like the short
females than like the No. 17 females (mean differences 4 X S.E. and
6XS.E.).
It was thought that the mean c.l./Ii.ic. ratio for short males from
experiment I data supplemented by the same ratio for short females
from stocks used for experiment II could be used in making com-
parisons with that ratio for biparental males of experiment II. Since
the small change in temperature appears to have had an unexpectedly
pronounced effect, and since male and female ratios are so different,
these comparisons can not well lie made. The biparental males of Ex-
periment II are, however, very significantly different from their haploid
(type wings) brothers with respect to mean c.l./h.w. ratio (Difference
--11 ;' S.E.). It is, thereby, clearly demonstrated that short was in-
herited from the father by these biparental males.
Data supplementary to that of Experiment I were obtained from
Dorothy A. Binns (Table VII. 11)1. The wasps measured by her were
reared under the same conditions as those of Experiment I.
The conclusion that biparental males are heterozygous for short
seems justified for three reasons. 1. In Experiment I biparental males
were shown to inherit short from their mother and in Experiment 1 1
they were definitely shown to inherit it from their father. 2. The mean
c.l.'h.io. ratio for biparental males of Experiment I is significantly dif-
ferent (4 X S.E.) from that of their short brothers and is also sig-
nificantly different (4 X S.K.) from that of the type (wings) males
(No. 17) of Experiment II. The difference from that of the type
males (No. 1) of Experiment I would be even greater since No. 17
probably carries minor factors for short. 3. ( hi the basis of supple-
mentary data (Table VII. 11» biparental males of Experiment I are
shown to be intermediate between and markedly different from the two
types of haploid males (normal and mutant) in mean c.l./li.u'. ratio.
DIPLOIDISM IN HABROBRACON 151
It may then be concluded from this study of inheritance of four
factors determining- distinguishable heterozygous types that biparental
males are duplex for these factors.
P). Tests of I>i parental Males
One of the strongest proofs of the diplnidy of biparental males is the
manner in which they breed. It was previously shown (Whiting, Anna
R., 1927 and 1928) that over seventy-five per cent of biparental males
were entirely sterile and that the remainder had produced a very small
number of daughters which showed, with but rare exceptions, the
dominant traits of the male parent. A few exceptional cases were re-
ported early in the work on Habrobracon (Whiting, P. W., 1921).
Seven biparental males with black eyes produced daughters with orange
eyes; these males, therefore, bred like mosaics. Another exceptional
male produced a fertile daughter with recessive eye color (Whiting,
Anna R., 1927). As suggested, this may have been a thelytokous
daughter of the female used for the test. One exceptional male was
found (Whiting, Anna R., 1927) to breed as a heterozygote.
Table VIII gives a summary of the tests of biparental males made in
connection with this study. Whenever possible, males from each type
of cross were tested by mating them to females homozygous for the
recessive factor involved. Individual males were sometimes mated to
different females on successive occasions. Since this was in part a test
of sterility, matings were always observed.
The data show that 90 per cent of the biparental males tested were
sterile. Occasionally, one of these males had abnormal abdominal
sclerites, abnormal legs, wings or antennae but most of them appeared to
be normal and all of them mated with females. There must, therefore,
have been something abnormal about the spermatozoa or spermato-
genesis of these males.
On the basis of tests here reported (Table VIII) we would expect
one out of ten biparental males to produce daughters. Only one
daughter was obtained from 90 matings for which 34 biparental (Sh sli]
males were used (Table VIII). Sixteen daughters were produced by
5 of the 12 biparental (Ta ta) males tested. Only 17 matings of Ta ta
males were observed and of these six were fertile. Four biparental SI si
males, out of 20 tested, were fertile. Fertility of biparental males seems
to differ with the factor involved and perhaps with the stock, though this
point has not been tested particularly and is merely indicated by the
above figures.
152
MAGXHILD M. TORVIK
TABLE YIN
Tests of Biparental Males
Composition of Males
No. of Males Tested
No. of Matings
Offspring;
Total
Fertile
Total
Fertile
9 9
tfd"
\]'wOo
17
9
5
14
12
3
20
12
2
2
1
22
16
14
2
38
34
2
1
0
3
5
0
4
0
0
0
0
4
1
3
0
2
1
33
9
5
16
17
3
35
17
2
3
1
29
28
16
5
56
90
2
1
0
3
6
0
4
0
0
0
0
5
1
-»
0
2
1
2
1
0
18
16
0
25
0
0
0
0
17
1
5
0
5
1
862
153
503
1189
1184
173
1264
687
47
134
2
904
373
677
517
2586
2685
Lit 1,1
Kikr)ksO(o)o' '*
Tata
Wawa
Slsl
F/0(o)o'' *
MniOo
BbOo
SpspOo.
CcOo(O) *
MamaOo(O) *
odoi.
Ststoo*
YyO(o)of *
ShshOto^o * .
Totals ....
223
26
365
28
91
13934
* Symbols set in parentheses may he substituted for immediately preceding
symbols.
C. Genetic Composition of Daughters of Biparental Males
Daughters of biparental males were found (Whiting, Anna 1\., 1925
and 1927) in previous experiments to he more nearly sterile and more
abnormal than the biparental males. Certain of the 91 females obtained
during the course of this investigation were also somewhat abnormal.
Some of them had wrinkled wings or abnormal antennae or irregular
abdominal sclerites and a few had irregular gonapophyses, but only 7
were not tested with cakTpillars. Seven died without stinging the cater-
pillars. Stung caterpillars were used with most of the others. Of the
77 thus tested 66 fed on the caterpillars but laid no eggs, 7 laid eggs
which dried up without developing and 3 produced offspring.
The 3 fertile daughters, type in appearance, all came from one of the
Ta ta males mentioned above. One daughter was produced in vial " a "
(the first vial) by a heterozygous Ta ta female which ran through " (/ '
(the fourth vial), producing also 58 males. This fertile daughter pro-
duced one type daughter which bred as a heterozygote, giving 10 taper-
ing and 1 1 type; males. From a second mating of this male with a
tapering female there were obtained in " a " 18 tapering males and 2
1HPLOIDISM IN HABROBRACON 153
sterile type females, in " b " 24 tapering males and 2 fertile type females
and in " c " 13 tapering males. One of the fertile females produced one
tapering male, the other produced 4 tapering males and 3 tapering
females.
Since stung caterpillars were used with the tapering female of the
second mating and for testing the daughters, contamination may explain
the results. However, no cases of contamination were noted in other
tests and 13 other daughters of Ta ta males were tested and found sterile.
The daughters of biparental males, in almost all cases, show only the
dominant traits of their male parent regardless of how these traits
entered his composition. In many cases one dominant was contributed
by the male parent and another by the female parent of the biparental
male and the daughter showed both of the traits.
It was interesting to note the appearance of characters determined bv
factors showing incomplete dominance. Daughters were obtained from
od-ol males mated to O-o' and o'o' females. The eye color of these
daughters was about the same as that of biparental males or hetero-
zygous females (o''-o'). Two Yy males produced daughters. One
male was very dark; it was mated to a type (antennae) female and
produced 4 daughters with rather dark antennae. The other was lighter ;
it was mated to a Yy female and produced a daughter having the 3 basal
antennal segments clear yellow.
The small eye locus gave interesting results. Eighteen daughters
were obtained from 3 matings. In 2 cases Kks males were mated to
type females and produced 14 daughters, of which 4 had eyes much
smaller than normal. In the other case, a Kks male was mated to a
small eye female and the 4 daughters resulting had small eyes — much
smaller than normal though not as small as small eye often is. This
locus needs to be tested further.
For other factors (Table VIII), where dominance is known to exist,
the daughters always showed the dominant traits of their male parent
though he must also have carried recessives. These males, then, do not
behave like diploid individuals ; they do not segregate recessives. In
order to explain this discrepancy it has been suggested that the sperm
may be diploid.
If the sperm of biparental males are diploid, their daughters may
be expected to be triploid. They must possess some irregularity in
chromosomal constitution since they are often morphologically abnormal
and rarely produce offspring.
The triploid nature of these daughters has been tested by means of
sets of three pairs of factors, each dominant being brought in by a dif-
ferent individual. Two sets of such factors have been used.
154 MAGNHILD M. TORVIK
In one instance the complementary eye color factors orange and
cantaloup were used along with the factor for reduced wings. Seven-
teen biparental males resulted from the union of egg (oCr) with sperm
(Ocr). These males had type (black) eyes and reduced wings. They
(OoCcrr) were mated to orange cantaloup females (ooccRR). The
16 daughters from this cross were type showing the three dominants
(OCR). This indicates probable triploidism.
More conclusive proof of the triploidy of the daughters was obtained
by means of two linked (Cc and LI) and one independently segregating
(Rr) pairs of factors. Cantaloup (eye color) and long (antenna,
wings and legs) are linked and have a cross-over value of about ten per
cent. Eighteen biparental males were obtained from crosses of canta-
loup reduced (cL/cL r/r) females by long reduced (Cl r) males. The
biparental males were type except for reduced wings and must, there-
fore, have been duplex for the cl chromosome. These reduced males
cL/Cl r/r) were mated with cantaloup long (cl /cl R/R) females.
One daughter was obtained. She has been thoroughly examined and
compared with long and reduced wasps. One of her primary wings was
not completely expanded, but camera lucida drawings were made of the
other and of her antenna? and legs. These were compared with draw-
ings of long. She was found to be type with respect to all three struc-
tures, showing that she possessed the dominant factor L. She was
obviously non-reduced, R, and non-cantaloup (black-eyed), C, and hence
must have resulted from union of sperm (cL/Cl r/r) with egg (cl R)
and would, therefore, be (cl/cL/Cl R/r/r) triploid.
Since daughters of biparental males are often morphologically ab-
normal, rarely produce offspring and will inherit from three individuals
in the manner shown above, it seems probable that they are triploid.
IV. DISCUSSION
Data thus far obtained suggest that biparental males are duplex for
all chromosomes studied. None of the seventeen mutations tested gave
evidence of being simplex in hiparental males and therefore located in
a sex chromosome. Genetically we have as yet no sex chromosome and
the question as to why biparental males are males is still unanswered.
Sex in Drosophila has been most conclusively shown to be dependent
upon genie balance. Even the haploid has recently been shown
(Bridges, C. B., 1930) to be fc-mak-.
However, the situation in Drosophila must differ somehow from that
in certain other forms where males are produced through haploid
parthenogenesis and females are diploid. This apparently is true of
most Hymenoptera (Whiting, P. \V., 1918) and has he-en definitely
D1PLOIDISM IN HABROBRACON 155
shown to be true of many coccids. In Iccrya purchasi it was found
(Schrader, Franz and Hughes- Schrader, Sally, 1926) that, "the indi-
vidual chromosomes of the haploid set as found in the cells of the males
correspond in size to the individual chromosomes of the diploid, — one
member of each of the two morphologically distinct pairs of the diploid
set apparently being present in the haploid group." The same is true of
other coccids (Hughes- Schrader, Sally, 1930).
Sex in such forms can hardly be based on the same sort of genie
balance as in Drosophila. Bridges (1925) suggested that, "at present
the difference between haploid and diploid sexes must be referred to the
same type of determination as that responsible for the larger size,
rougher texture of eyes and other slight changes that distinguish the 3 N
from the 2 N individual," in Drosophila.
Castle (1930) postulates that there is no minus sex-tendency in
species which have haploid males. ' The egg is homozygous for plus
sex tendency (XX) ; the haploid male transmits in its one class of
sperm this same sex-tendency (X)." He is unable to state why these
haploid individuals, genetically female, are phenotypically male.
One interesting point is the fact that there has been no indication of
intersexuality in Habrobracon juglandls (Ashmead) though cases of
intersexuality were noted in H. brcvicornis (Wesmael) (Whiting, P. W.,
and Whiting, Anna R., 1927, p. 112). The more biparental males are
studied, the more closely is their genetic composition found to parallel
that of the diploid female. In spite of this genetic similarity to females
the biparental males are as definitely male in morphology and reactions
as are haploid males nor does their added chromatin make them notice-
ably larger than haploid males.
V. SUMMARY AND CONCLUSIONS
1. Biparental males inherited dominant allelomorphs, of eleven pairs
of factors showing ordinary dominance and recessiveness, from both
parents when reciprocal crosses could be made or from either parent
carrying the dominant when they could not.
2. Biparental males inherited complementary dominant factors, one
from each parent, in such a way as to reconstitute the type character.
3. Homologous chromosomes were shown to be present in biparental
males by means of linked factors.
4. Four factor pairs which produce distinct heterozygous types were
inherited by biparental males in such a way as to demonstrate a duplex
condition in each case.
5. Daughters of biparental males were shown to inherit, in almost
all cases, only the dominant traits of the male parent regardless of how
these traits entered his composition.
156 MAGXH1I.D M. TORVIK
6. The triploid nature of these daughters and the diploid nature of
the sperm of biparental males were tested by means of three pairs of
factors, each dominant being brought in by a different individual.
7. Biparental males appear to be- diploid and their daughters appear
to lie iriploid.
LITERATURE CITED
oES, C. B., 1925. Sex in Relation to Chromosomes and Genes. Am. Xut..
59: 127.
BRIDGES, C. B., 1930. Haploid Drosophila and the Theory of Genie Balance.
Science. 72: 405.
CASTLE, W. E., 1930. The Quantitative Theory of Sex and the Genetic Character
of Haploid Males. Proc. Nut. A cod. Sci.. 16: 783.
DZIERZOX. J. EICHSTAED, 1845. Bicucn, Ztg. 1.
Hi'GHES-ScHRADER, SALLY, 1930. Contributions to the Life History of the Iceryine
Coccids, with special reference to Parthenogenesis and Hermaphroditism.
Ann. Entom. Soc. America, 23: 359.
SCHRAHKR, I-'KAXX AND HuGHES-ScHRADER, SALLY, 1926. Haploidy in Iccrya
purchasi. Zeitschr. f. u'iss. ZoiU.. 128: 182.
TORVIK. M. M.. 1929. Are Habrohracon Males Diploid for the X-ray Mutation
" Short "? I' roc. Pom. Actnl. Sci., 3: 2.
WHITING, A. R., 1925. The Inheritance of Sterility and of other Defects Induced
by Abnormal Fertilization in the Parasitic Wasp, I labrobracmi juglandis
(Ashmcad). Genetics, 10: 33.
WIIITIXG, A. R.. 1927. Genetic 1-lvidence for Diploid Males in Habrobracon.
Biol. Hull., 53: 438.
WHITIXG. A. R.. 1928. Genetic K\i<lrnce for Diploid Male> in Habrobracon.
Am. Nat., 62: 55.
WHITIXG. P. W., 1918. Sex Determination and Biology of a Parasitic Wasp,
1 ladroliracoii brevicornis (Wesmael). Biol: Bull.. 34: 250.
WHITIXG, P. \\'.. 1921. Sludie> on the Parasitic Wasp, Hadrobracon brevicornis
(Wc-smael). I. Genetic^ of an orange-eyed mutation and the production
of mosaic males from fertilized eggs. Hiol. Bull.. 41: 42.
WIIITIXG. P. W., AXD \\'IIITIXG, A. 1\.. 1927. Gynandromorphs and other Irregu-
lar Types in Habn.bracon. Biol.'Bnll.. 52: 89.
THE CHROMOSOMES OF Till-: DOMESTIC TURKEY
ORILLA STOTLKU WKRNER
COTTKY JUNIOR COLI.E<;E VOR \\'OMK\, NEVADA, MISSOURI
In a previous article (Biological Ihilletin, Vol. LIT, No. 5. May,
1927), I have described the chromosomes of the Indian runner duck,
giving the probable number and forms, and have proposed a scheme for
sex-linkage and sex-determination. The present study was undertaken
in 1927 to find out whether or not the conditions found in the duck
exist in other avian forms.
This study is based on the examination of approximately 800 mitotic
figures taken from 35 individuals. The same four general methods of
technique were employed as in the former work, except that in the tech-
nique for sectioned material of the testis, Bergamot oil was used instead
of cedar oil. Since the tissues of the turkey were more difficult to
prepare than those of the duck, the utmost precision was necessary in
order to obtain desired results.
As in the cluck, the cells of the male contain an even number of
chromosomes, and the cells of the female an odd number and one more
than is found in the cells of the male. The number of chromosomes
in the turkey appears to be the same as in the duck.
In the embryonic tissues of the male one cell was found which
appeared to have 66 chromosomes (Fig. 2), but a large majority of the
cells examined contained 76 chromosomes. In the embryonic tissues
of the female two cells were found which contained 55 chromosomes
each. One of these is shown in Fig. 7. The remainder of the female
cells examined appeared to contain 77 chromosomes each. It would
appear that the typical somatic numbers are 76 for the male and 77 for
the female.
As in the duck the chromosomes are of three forms: T-shaped, rod-
shaped, and globe-shaped, and as in the duck the chromosomes fall into
three general groups. In the male of the turkey these groups consist
of 6 pairs of large chromosomes, 3 pairs of short rod-shaped chromo-
somes, and 29 pairs of globe-shaped chromosomes. In the female there
are 6 pairs of large chromosomes, resembling in si/e and form those of
the male, plus one odd chromosome which is the largest in the group ;
3 pairs of short rod-shaped chromosomes; and 29 pairs of globe-shaped
chromosomes. It is apparent that the difference in the chromosomal
157
158 ORILLA STOTLER WERNER
grouping in the cluck and in the turkey is in the second and in the third
groups. In the duck the second group contains nine pairs of short rod-
shaped chromosomes and the third group contains 23 pairs of globe-
shaped chromosomes.
In the aberrent cells of the male which contain less than 77 chromo-
somes (Fig. 2) all of the twelve large chromosomes of the first group
are present, also the six short rod-shaped chromosomes of the second
LToup. The missing chromosomes are the ten smallest of the third
group. The same thing seems to be true in the aberrent cells of the
female of less than 77 chromosomes. In these are present the 13
large chromosomes of the first group, the six rod-shaped chromosomes
of the second group ; but 22 of the smallest chromosomes of the third
group are not present.
The same pliancy is noted in the chromosomes of the turkey as was
evinced in the chromosomes of the duck. Because of this the chromo-
EXPLANATION OF PLATES
All figures from the turkey are reproduced at the same scale. The drawings
were outlined with an Abbe camera lucida at a magnification of 3,500 diameters,
obtained with a Spencer 1/12 homogeneous immersion objective and Spencer 15X
compensating ocular with draw tube set at 150 mm., and drawing made at the level
of the base of the microscope. The drawings were then enlarged by means of a
copying camera lucida to 7,350 diameters. Having been reduced one-third in the
reproduction they now appear at a magnification of 2,450 diameters.
W , the large sex-chromosome carrying female-tendency genes only.
u>, the smaller sex-chromosome which also carries female-tendency genes only.
Z, sex-chromosome carrying a preponderance of male-tendency genes and
also sex-linked genes.
38Z, same as Z.
38?c, same as u>.
37 to 1, autosomes.
1 EXPLANATION OF PLATE 1
FIGS. 1—4. Cells from the amnion of males of the domestic turkey. The sex-
chromosome is numbered 38Z. Autosomes from 37 to 33 are paired according to
their size.
FIG. 1. Early prophase. The large chromosomes have not yet taken the
characteristic peripheral position. Seventy-six chromosomes present.
FIG. 2. A soma cell aberrent in chromosomal number. This cell has the
12 large chromosomes of the first group, the 6 rod-shaped chromosomes of the
second group, and 48 chromosomes of the third gmnp. Ten of the smallest
chromosomes of the third group arc missing.
FIG. 3. Early metaphase showing gonomeric grouping of the largest chromo-
somes. Seventy-six chromosomes present.
FIG. 4. Early metaphase showing gonomeric grouping of the largest chromo-
somes, also some filamentous linkage between the members of the third group.
Autosomes from 37 to 33 numbered. Seventy-six chromosomes present.
FIGS. 5 and 6. First spermatocytes from smear preparations. These cells arc
in the prophase stage. The Z chromosome bivalent in each cell is numbered
38Z, the autosomal bivalents from 37 to 1 according to their size. Thirty-eight
chromosomes present.
CHROMOSOMES OF THE DOMESTIC TURKEY 159
PLATE I
v •<>.*
fcl •£*
Fig. 1
Fig. 3
o
-Js
33
o
Fig. 2
Fig. 4
Fig. 5
Fig. 6
160 ORILLA STOTLER WERNER
somes are often found in more or less modified forms at the beginning
i if mitosis.
The seventy-six chromosomes in the complex of the soma cells of the
male (Figs. 1, 3, 4) appear to form a graduated series from the largest
of the first group, 3SZ, to the smallest of the third group. Of the
twelve large chromosomes which form the first group numbers 38Z, 37
and 36 appear in most cases as J-shaped and 35, 34 and 33 as rod-
shaped. The six chromosomes which form the second group are all
of the rod type. They are in most cases sufficiently smaller than the
smallest of the first group so as not to be easily confused with them, but
the slight difference in the length of the individual members of the three
pairs makes it sometimes difficult to distinguish one from the other.
The third group, the globe-shaped chromosomes, range in size from
those containing approximately as much chromatin as the smallest of
the second group to very small ones. Many of these small chromosomes
are so nearly of the same size that they can be compared with no degree
of certainty. Figures 5 and 6 are first spermatocytes from smear
preparations of testis material. The tetrad form of manv of the
chromosomes is plainly apparent. It would appear that gonial mates
have the same spindle attachment and that they are of the same size.
In these cells the haploid number i^ present and the grouping of the
chromosomes and the size relations are the same as in the diploid num-
ber. The germ cells are so small that in sectioned material it is prac-
tically impossible to be sure of the small chromosomes; but in most cases
the large chromosomes are easily made out as to number and form.
However, the testis material lends itself well to smear preparations and
when the cells are well pressed out. the chromosomes are sufficiently
large and clear to distinguish their number and form.
EXPLANATION OF PLATE ! I
FIGS. 7 to 12. Cells from the amnion of fi-malcs of the domestic turkey.
/(' and 387C1 are the chromosomes that carry female-tendency factors only. Auto-
somes paired from 37 to 33.
FIG. 7. A pro])hase aluTmit in chromosomal number. There arc present the
thirteen large chromosomes of the first group, the six rod-shaped chromosomes of
the second group, and thirty-six of the chromosomes of the third group. Twenty-
two of the smallest chromosomes of the third group are missing. Fifty-five
chromosomes present.
FIG. 8. Prophasc. Seventy-seven chromosomes present. Some filamentous
linkage shown. Seventy-seven chromosomes present.
FIGS. 9 to 12. Cells in metaphase. The stippled line in each case shows a
possible gonomeric grouping. This grouping in each case has heen considered with
especial reference to the twelve largest chromosomes. Filamentous linkage of
chromosomes shown in eacli cell. Autosomes paired fnun 37 to 33. Seventy-
seven chromosomes present.
CHROMOSOMES OF TIIK DOMESTIC TURKEY
PL ATI- II
161
Fig. 7
Fig. 8
Fig. 10
Fig. 9
Fig. 11
Fig. 12
162 ORILLA STOTLER WERNER
As in the duck there is one more chromosome in the complex of the
female than in the male. There are probably seventy-seven (Figs.
8-12). The odd chromosome is the largest in the complex. In these
figures it is designated as U\ This large chromosome, which is evi-
dently a sex-chromosome, is quite pliant and apparently adjusts itself
t» other chromosomal regions and to the nuclear wall. For some time
this chromosome was regarded as rod-shaped, but in anaphase it is
J-shaped. However, it differs from the other J-shaped chromosomes
in the complex in that the end that forms the loop of the / is tapering
whereas the other J-shaped chromosomes are approximately uniform
in diameter throughout their entire length. The next two largest
chromosomes (3Sw-38Z) are apparently J-shaped. As in the duck it
is difficult to say whether or not they are gonial mates. If the theory
advanced in the former study concerning sex determination is correct
then they are not gonial mates, but one of them must be regarded as a
homologue of the 38's in the cells of the male and the other as a w
chromosome. The remaining chromosomes in the complex of the
fem.nle appear approximately the same as those of the same numbers
in the cells of the male.
As in the duck gonomeric grouping is evident. In the metaphase
of males (Figs. 3, 4) and in the metaphase of females (Figs. 9-12).
As in the duck "... in the cells of the male the largest chromosomes
are grouped six on one side of the forming equatorial plate and six on
the other. ... In the cells of the female there are six on one side of
the plate and seven on the other. ... In every case there is in the group
of seven, one chromosome which is larger than the others which has
the characteristic form of the largest odd chromosome in the cells of the
female, large at one end and taper at the other." i Werner. I'.ioln^ical
UulU-tin. Vol. LI I. No. 5, May, 1927).
I have not been able to determine whether the two halves of a nucleus
are of exactly the same size. It would appear that in some cases there
is some discrepancy in this respect. Neither have I been able to deter-
mine whether or not homologous chromosomes are of exactly the same
size in any one stage of mitosis. In pairing the chromosomes I have
selected as homologues those that are more nearly of the same shape
and length. It is possible that there is a difference in the amount of
chromatin material in some or all of the homologues. The chromo-
somes of the male may contain more than those of the female or vice
versa. This is a difficult question but one that should be investigated.
As in the duck filamentous linkage occurs in the somatic cells. In
most cases observed the linkage is between members of the third group,
the globe-shaped chromosomes. (Figs. 1, 4, 8, 9, 10, 11. 12.) In
CHROMOSOMES OF THE DOMESTIC TURKEY 163
other cases it was between members of the second group and some one
member of the third group. (Figs. 1, 11). The numbers of chromo-
somes thus attached in linear arrangement range from two to seven.
The filaments are in most cases one in number, although there are some-
times two. (Fig. 1). They are in all cases oxyphylic in character and
are somewhat roughened or crinkly.
DISCUSSION
The similarities between the chromosomal complexes of the duck
and of the turkey are at once apparent. In each form there is in the
male an even number of chromosomes, while in the female there is an
odd number of chromosomes and one more than is present in the cells
of the male. The same condition is found in the chicken (now being
investigated). As in the duck it seems probable that the largest pair
of chromosomes in the male complex are the Z or the sex-chromosomes.
Since the large ]V chromosome has been found in the female of the three
forms of the aves, it seems impossible to regard it as a planosome, or
supernumerary. It must, then, be regarded as an odd chromosome and
if such, it is reasonable to suppose that it is a sex-chromosome. Since
it is found only in the cells of the female, it is evident that it is concerned
only with femaleness. It is equally evident that it does not carry sex-
linked characters but that this must be the function of some other
chromosome in the female complex. The scheme proposed in the
former article for sex-linkage and sex-determination in the duck is
entirely applicable in the case of the turkey and it seems unnecessary
to repeat it in detail in this article. The generalities are that the female
tendencies are carried by the ]Vw chromosomes, the male by the Z
chromosomes. In both sexes the sex-linked tendencies are carried by
the Z chromosomes. The autosomes are in a balanced condition be-
tween maleness and femaleness. It follows that a zygote receiving
a genie complex equally balanced between maleness and femaleness, plus
that which contains genes for maleness only (the Z chromosome of the
male) would of necessity become a male. A zygote receiving a genie
complex equally balanced between maleness and femaleness plus the
Z chromosome, which contains genes for maleness only, and in addition
the IV iv chromosomes which carry genes for femaleness only, would
become a female. It would, of course, follow that the F^ and F.2 gen-
erations would inherit as is usual in such sex-linkage and as has been
outlined in the previous article.
164 ORILLA STOTLER WERNER
SUMMARY
1. The chromosomes in the somatic cells of the turkey agree in num-
ber with the chromosomes in the somatic cells of the duck. These
"appear to he 76 chromosomes for the male and 77 chromosomes for
the female. There is present in the cells of the female a long impaired
chromosome which is not found in the cells of the male. There is
reasnii to suppose that there are prohably among the remaining six
largest chromosomes two more unpaired chromosomes, one of which,
the largest, is probably homologous to the largest pair (sex-linkage) of
chromosomes in the male complex, while the other, it is thought may be
some one of the five remaining long chromosomes." (Werner.)
2. As in the duck the 76 chromosomes appear to fall into three
general groups, hi the duck these groups consist of six pairs of large
chromosomes, including three J -shaped and three rod-shaped; nine pairs
of short rod-shaped chromosomes; and twenty-three pairs of globe-
shaped chromosomes. In the turkey the first group consists of six pairs
of large chromosomes, including four pairs of J-shape and two pairs of
rod-shape. The second group consists of three pairs of short rod-shaped
chro'iiosi imcs ; the third group consists of 29 pairs of globe-shaped
chromosomes, which as in the duck, form a closely graduated series.
3. As in the duck there appear to be 38 bivalents in the primary
spermatocytes of the male. These agree with the somatic cells in size
gradations.
4. Gonomeric grouping occurs in the amnion cells as it does in the
duck.
5. Filamentous linkage occurs in certain stages of the prophase and
metaphase. This also agrees with the condition found in the duck.
6. The sex-mechanism appears to be of the //\v/-ZZ type similar
to that found in the duck and in the moth Phragmatobia.
I desire to re-cord my indebtedness to Dr. \Y. K. l'>. Robertson for
the material for this work and for his criticism of the major part of
the work; to the ]>aiiM-h and I.omb < >ptiral Company for the use of
microscope equipment during the year 1(>29; to Dr. Mary Rose Prosser,
President of Cottey College, and to Mrs. Elizabeth Ott for their in-
fluence in securing from the Spencer Lens Company the proper equip-
ment for the completion of the work.
THE SIGNIFICANCE OF HYDROGEN ION CONCENTRA-
TION" IX THK BIOLOGY OF KfGLENA GRACILIS
KLEBS
GORDON ALEXANDER
(From the Physiological Laboratory, Princeton University)
INTRODUCTION
Eiiglcna rjracilis Klebs is a common and important constituent of
certain aquatic communities, but is easily cultured under laboratory
conditions. Hence, it is especially suitable for investigations in the
ecology of a single species, and the present study is a contribution to
that field. As Allee (1930) pointed out in his presidential address
before the American Society of Ecologists, we do not begin to know as
much about the morphology and physiology of individual species as is
desirable, and the mere cataloguing of organisms from different en-
vironments has yielded little of real value.
The present study is designed to show the effects of different H+-ion
concentrations in the external medium on Euglcna gracilis, with other
factors controlled in such a way that indirect effects, or effects from
unknown variables, are reduced to a minimum. In nature, probably
most of the effects of H+-ions are indirect, but we can discover their
true nature only by eliminating them under controlled conditions. For
this reason, in the present studies, cultures free from all other organ-
isms have been used.
Certain aspects of the physiology of Euglena gracilis are fairly well
known, due to the researches of Klebs (1883), Zumstein (1899),
Ternetz (1912) and others. In particular, its tolerance of high con-
centrations of citric and other acids was pointed out by Zumstein and
Ternetz, both of whom made use of citric acid in the more or less
complete elimination of bacteria from cultures of the Euglcna. Kostir
(1921) demonstrated that this high degree of tolerance for citric acid
is, however, a species characteristic not generally true for the genus.
Zumstein emphasized the necessity of using bacteria-free cultures,
asserting that the presence of bacteria materially depressed the division-
rate. He believed that the increased division-rate in acid cultures was
not a direct effect of the acid, but an indirect effect, through the elimina-
tion of the bacteria. His evidence was incomplete, but my own experi-
ments demonstrate that his view was correct.
165
12
166 GORDON ALEXANDKR
It is possible that other organisms may be similarly effective.
Skadowsky (1926). studying cultures of mixed Protozoa in relation to
H^-icn concentration, arrived at an optimum pi I for lliiglcna gracllis
Klcl>- of about 3.8. a value very different from that which I have found
in pure-line sterile cultures but not far from my finding in a series
nf cultures in which bacteria were present. Other Protozoa in
Skadowsky's cultures showed different pi 1 optima, his results indicating,
for those species listed, almost no competition at optimum values.
l:n(/!cna gracills actually grows well in a wide pH range, as will be
shown later, with a not very pronounced optimum. Therefore, may
not this apparent (and very striking) optimum found by Skadowsky
be due to the competition of one or more other forms at the true
optimum for this species, depre-.ssing the division-rate of the linglcna
below that at which it lias no competition? Interspecies competition is
very real, certainly applying to Protozoa as well as higher forms; and,
in a case like that in question, it may mask the real responses of the
individual species. The results obtained by Skadowsky may well apply,
therefore, to l-.uglena gracllis in the community which he studied, but
not to this species when isolated from others. The responses under the
later conditions are fundamental to the particular species considered, but
subject to modification bv the presence of other forms. In studying
the responses of an organism to an environmental factor under natural
conditions, one must remember that the results of such findings apply
only to the special complex community in which the study is carried out.
As far as the writer is aware, no observations of the present nature
have previously been carried out on any of the luiglenoidina. A few
related studies on the alga, Clilorclla, have been made by Warburg
(1919) and \Yann and Hopkins (1927). A verv good summary of
studies on pi I in relation to Protozoa, ciliates in particular, is that of
Darby (1929). whose observations arc extended in a later paper (1930).
The earlier paper contains a good bibliography. A recent paper of
general interest, which empha>i/.es the significance of CO.. in influencing
the pH of natural waters, is that of Powers (1930). A most satis-
factory summary of the relations between pi I and fresh-water and
marine organisms is that hv P>rcsslau (1926). A long bibliography is
appended. Skadowsky's paper is of similar general interest.
Thanks for special favors in connection with the present study are
due to Dr. \\ . I*. I'aker. of Kmory University, for verification of my
identification of the first specimens used; to Dr. ('. II. Philpott, of
Harris Teachers' College, for a demonstration of the technique ot
sterilizing /'aniniccimii : and to the Digestive Ferments Company, of
Detroit, Michigan, for information connected with the analysis of their
SIGNIFICANCE OF pH IN FUGLENA
167
product, ' Bacto-peptone." To Professor E. Newton Harvey, I am
especially obligated for continuous advice and assistance in technical
aspects of the problem, to say nothing of the inspiration gained from
association with him in his laboratories.
EXPERIMENTAL Al KTIIODS
The apparatus and methods of procedure used were, of course,
necessitated by the desire to maintain unvarying from culture to culture
all important factors, except pH, in the growth of Euglcna gracilis.
These factors are not only those affecting growth or reproduction, but
also the limiting factors for photosynthesis. (Blackmail, 1905; Stiles,
1925; Spoehr. 1926.)
The following physico-chemical conditions were subject to control:
Temperature ; frequency and intensity of light ; a sufficient quantity and
quality of food for both saprophytic and holophytic nutrition; in
certain cases, carbon dioxide and oxygen tensions; hydrogen ion con-
centration.
At O TOR
FIG. 1. Diagram of apparatus used in experiments.
Biological factors considered were: Absence of (1) a complicated
life-cycle, (2) racial differences, and (3) other organisms; uniform
density of organisms in all cultures at the beginning of an experiment ;
uniformity in organisms used for inoculation, — secured by using stock
cultures of the same age and pH (in all but one experiment with
etiolated Euglcna ) .
For temperature control a water thermostat was adopted. This is
illustrated in Fig. 1. The container is a metal tank, all inside parts
168 GORDON ALEXANDER
painted black. The temperature balance is maintained between a loop
of copper pipe through which tap-water flows, and a knife heater. (A
significant aim unit of heat is also derived from the illuminating source.)
The regulation is by means of a Thyratron tube in circuit with a toluol-
mercury thermo-regulator, as described by Loomis, Harvey and MacRae
i 1930). Temperature fluctuations, as determined with a Beckman
thermometer, are less than 0.01° C.
As a source of light a 60-watt Mazda Daylight Lamp (frosted) was
used. This was rotated about its vertical axis (at about 150 revolutions
per minute) to provide uniform total distribution of radiation in all
directions. However, since experiments with a 40- watt lamp resulted
in rates of division as high as those with the 60-watt, the light intensity
used was not limiting photosynthesis. As controls for the detection of
photosynthesis and other light effects, identical cultures in absolute dark-
ness were maintained in all experiments. The culture-tubes were
covered with several coats of black varnish, and, further, separated from
tlie source of light by an opaque metal screen.
In the tank the water was kept in constant circulation to maintain
uniformity of temperature. The lamp was not suspended directly in
the water, but inside a Pyrex cylinder closed below and open above (a
beaker was used). Much of the heat from the lamp was conducted
away in the air. Between this Pyrex wall and the cultures (in Pyrex
test-tubes 18 X 150 mm.,) was a water thickness of 5.2 centimeters.
The total distance from the axis of rotation of the lamp to the center
of each test-tube was 12 centimeters. Distilled water was used in the
tank, to avoid deposition of films of carbonate on the glassware. The
test-tube rack is of aluminum, painted black.
Temperature is known to have a marked positive effect on rates of
photosynthesis; but death from high temperature is accelerated in high
concentrations of 11 -ions, as lias been shown by Oialklcy (1930) for
Paramcciwn. In a preliminary experiment with I-'-utjIcna gracilis I
have observed the latter phenomenon at temperatures as low as 35° C.
In the selection of a suitable temperature for the experiments one must,
then-f^re. compromise. I have actually used temperatures of 29° and
29.5° C.
Bacto-peptone, ' Difco " Standardized, was selected as the basic
culture medium. Previous workers have used "peptone" chiefly, a
fact which makes my observations the more comparable with theirs.
Furthermore, this medium is not only sufficient for saprophytic nutri-
tion, but also permits photosynthesis without the addition of other media,
—at least between pi I 8.5 and 3.5. In addition, Bacto-peptone is in
itself a very well buffered system, and is therefore especially suitable
SIGNIFICANCE OF pH IN EUGLENA
169
to\
\
0.5
O
0.5 /.O
cc. % HCJ
f.5 £.0
2.5
5.0
FIG. 2. Buffer curve for 1 per cent Bacto-Peptone in distilled water.
The original volume, to which were added the indicated quantities of
NaOH and HC1, was 10 cc.
for studies of H+-ion concentration (Fig. 2). Even within the pH
range at which the buffer action is least, the pH changes produced by
the growth in it of Euglcna are, during the periods of experiments, not
great (Table I).
The concentration of peptone used — dissolved in glass-distilled water
— has been 1.0 per cent. With the density of organisms studied, no
TABLE I
Change in pH of cultures during an experiment lasting 48 hours.
These figures are from the experiment plotted in Fig. 5.
Final pH
iiniicti pn
In Light
In Darkness
8.10
8.13
8.07
7.65
7.72
7.70
6.71
6.87
6.87
5.90
(..11
6.15
4.60
4.64
4.84
3.56
3.55
3.55
2.96
2.81
2.89
2.52
2.39
2.48
170 GORDON ALKXAXDKR
greater growth is observed in this concentration than in 0.5 per cent, but
its ImiTer value is somewhat greater.
The' diffcrc'iit pH values studied were produced by additions of hy-
drochloric acid, an acid probably completely dissociated, and which
hardly, if at all. penetrates living cells. Its anion. furthermore, is
such a common constituent, of living matter as to be a relatively insig-
nificant factor. The culture medium was always freshly prepared, and
autoclaved but a single time for twenty minutes at fifteen pounds
pressure.
This peptone is free from carbonates or bicarbonates in detectable
quantity, and as a source of COX for the organism it furnishes no more
than that produced by the oxidation of the food-stuffs of which it con-
sists. The medium is, therefore, free from a source of CO.. that might
be released with increased Hf-ion concentration. In any case, with
peptone as the culture medium, the experimental results indicate that
the major growth of llnylena gracilis — even in the light — is not due to
photosynthesis, and is. therefore, not dependent on a supply of carbon
dioxide.
The oxygen-tension, while apparently not a significant factor in the
decomposition of CO.,,1 is a verv important factor in saprophytic nutri-
tion. In cultures sealed from the air, the photosynthesis of the con-
tained organisms provided the oxygen supply. — a supply which varied
with the extent of photosynthesis. The greater the rate of photosyn-
thesis, the greater the oxygen production, and, presumably, the greater
the growth due to the oxidation of the organic foods in the medium.
In aerated cultures, on the other hand, the organisms always had avail-
able a supply of oxvs^en (as well as carbon dioxide) in more or le~-
complete equilibrium with the air. In cultures only initially aerated,
and with access to the air at the surface, growth was somewhat limited
by the oxygen-tension; but. in several experiments, a constant satura-
tion with air was maintained.
The only factor intentionally varied from culture to culture has been
the H*-ion concentration. The range studied with most care has been
between the limits of about pll S.5 and 2.4. but the absolute limits of
life have been approximately determined.
For measuring H'-ion concentration, the quinhydrone electrode
method was adopted. It is dependable to 0.02 pll. or less, below a
pH of about 8.5. and is quite as convenient and rapid in use as the
colorimetric method, — at least for pigmented solutions. By this method
1 ll.irvcy (1(>2S) lins shown that decomposition of CO- can take place in
absence of oxygen. The method which lie used with marine alg:c, i.e., luminous
bacteria, as the indicator, was applied to liiiiilrnn f/racilis with results similar ;o
those obtained with the algae.
S1CX1FICANCK OK pll IX HUGI.KXA 171
it is possible to determine with accuracy such pi I changes as occur in
an active culture during the period of an experiment, the direction of
change giving a clue to the occurrence or absence of photosynthesis.
The values were recorded to 0.01 pH.
The stock cultures were grown in peptone medium unmodified by
the addition of acid (pll between 7.2 and 7.9), with continuous artificial
illumination, at room temperature. The chlorophyl content of the or-
ganisms did not vary materially from culture to culture. One study,
however, was made with etiolated Ititi/lciui from a stock culture of a
lower pH. The differences in division-rate between the green and
etiolated ones were not great. Hence, slight differences in the
chlorophyl content could not make appreciable differences in the results
—except when using inorganic media alone.
Because of the so-called allelocatalytic effects of Robertson (1922),
or a decreasing division-rate with increasing numbers ( Jahn, 1929), it
is necessary to compare cultures which begin with approximately the
same numbers of organisms. By using for inoculation stock cultures
that, to the eye, appear to be of about the same density, one can ap-
proach this condition. In any one experiment, each culture was inocu-
lated with the same volume of stock culture. The variation in numbers
per unit volume, as determined by actual counting in every case, was
less than 10 per cent.
For counting, I have used a Rafter Counting Chamber, 1.0 cc. ca-
pacity, with a squared disc in the microscope ocular. Before counting,
the organisms were killed and allowed to settle on the bottom. They
were killed by heating to 65 n C, a temperature at which they are
coagulated but do not disintegrate under any of the observed conditions.
The counts by this method were consistent to under 10 per cent (totals
of ten counts each, from different parts of the chamber), even in cul-
tures of lowest density of population. The error in the initial count
probably determines the total error in counting. It is not greater than
10 per cent.
The chief source of error in the present experiments is in this deter-
mination of the numbers of individuals at the beginning and end of
"runs." Slight differences in temperature, light. C( X-tension, and
other physico-chemical factors, would probably not produce variations
greater than those between different counts of the same culture. The
possibility of deviation from the true results is 10 per cent on the basis
of counts alone, this probably determining the error in the experiments.
We are not justified, therefore, in emphasizing any differences between
cultures unless they are of a greater order of magnitude than about ten
per cent — though consistent differences of less than that are probably
real.
17_> GORDON ALEXANDER
Genetic variables have been eliminated as far as possible by the use
of organisms all descended from a single ancestor — a pure-line or clone.
Variation clue to complicated life-cycles has not been demonstrated in
Eiti/lciur. (Reports of conjugation have been received with much
scepticism.) Encystment occurs, but it was not observe! in the course
of any of my studies on pH. I found cysts in old stock cultures, of
course, but none in the test-tube cultures after the two, three or four
days of my experiments. Encystment is, therefore, not induced by any
pH within the range 8.5-2.4. This is rather interesting, for encystment
(if some ciliates seems to be related in pi I nxoffmann, 1924; Darby,
1929). I have found no evidence for a regular cycle involving encyst-
ment in Jluglena gracills, nor am I aware that any previous workers with
this species have found such a cycle. Reproduction is, according to my
observations, due solely to longitudinal fission. In the present study,
the rate of reproduction is taken as the chief criterion of the effects of
pH. This rate has been determined by considering the initial and final
counts in a culture the first and last terms of a geometric progression
with two as the common ratio.
The method used in establishing a sterile, pure-line culture was that
used by Hargitt and Fray (1917) and Philpott (1928) for Paramecium.
The parent individual for the present experiments was isolated October
2, 1930 and washed by transfer from one to another of five drops of
sterile medium, being left in each for two or three minutes.2 This
individual and others sterili/ed at the same time were separately trans-
ferred to test-tubes, each containing about twenty cubic centimeters of
sterile medium. Each culture began to. appear greenish within about
a week. Absence of other organisms was verified by making plate
cultures on nutrient agar. The Knylcua appeared on the agar, and
reproduced rapidly, but no other forms were observed.
Experimental cultures were maintained sterile by ordinary methods
of bacteriological procedure. Cultures sealed from the air were closed
with sterilized, paraffined corks, tinfoil covers for corks having proved
toxic in preliminary experiments. Air bubbled through sterile cultures
was always sterili/ed by first passing it through sterile cotton, as illus-
trated in Fig. 1 .
RESULTS
Preliminary studies demonstrated that the previously reported in-
crease in numbers of Euglcna gracilis with increased concentrations of
citric acid was largely related to the II -ion concentration. Maximum
2 Parpart's criticism (1928) of the method of Hargitt and Fray does not
apply in the case of Euylcna, since it is extremely doubtful if Eitylcna ever ingests
solid food.
SIGNIFICANCE OF pll IN EUGLENA 173
growth in cultures made up with either hydrochloric, sulphuric, oxalic
or citric acid occurred at pH 3 to 4, if no sterile precautions were
observed. At higher pH values bacteria were numerous, especially near
neutrality, but at the "optimum" few were present. Media adjusted
to pH 3.6-3.7 by valeric or salicylic acids were fatal to Euglcna gracilis,
although the organisms thrived in more acid cultures prepared with the
other four acids. The easily penetrating acids, as one should expect,
proved fatal. The effect of the other acids was, however, obviously
related to H+-ion concentration, being independent of the kind of acid
used. Subsequent experiments have shown that the increased growth
at low pH is not a real effect of pH, but is an indirect result of the
elimination of bacterial competition at these values.
In studies on a sterile pure-line, carried out in 1925, the maximum
division-rate when CCX was the only acid added occurred at about the
same pH of the medium as when hydrochloric acid is used, — viz., pH
6.5 to 6.9. Concentrations of CO2 above that represented by 5 per cent
saturation at room temperature were not accompanied by increased
division-rates under the conditions of the experiments. From these
early observations I saw the desirability of, first of all, determining the
relations between pH, as such, and the rate of reproduction of Euglena
gracilis.
Scaled Cultures
Typical results from a series of sealed cultures are plotted in Fig. 3.
Not much weight can be placed on the irregularities in the figures from
the controls in darkness, but we can say that no striking effects are ob-
servable. The exhaustion of the limited oxygen supply, and the im-
possibility of its replenishment, make continued growth impossible at
any pH in the dark.
An interesting collateral experiment was undertaken in this connec-
tion to determine whether or not this Euglcna can live under anaerobic
or near-anaerobic conditions. Similar cultures, in tubes with capillary
necks, were exhausted as completely as possible with a vacuum pump.
Each culture tube was then sealed at the neck in a flame. One set was
placed in darkness, the other, in the light. Fairly rapid reproduction
occurred in the latter. In the culture in darkness, however, the organ-
isms were all motionless within twenty-four hours, and within another
day were all encysted. They died in the encysted condition, for, when
air was readmitted by breaking the sealed tip they did not excyst, nor
did they excyst when transferred to fresh medium.
These results may be explained in the following way : In both cul-
tures traces of oxygen and carbon dioxide remained after the exhaustion.
In the light, a trace of the latter would be sufficient to initiate photo-
174
GORDON ALEXANDER
synthesis. This process once begun, the linglcna would be self-suf-
ficient for both oxygen and carbon dioxide. But in darkness, after the
little oxygen present had been used, the organisms were in complete
absence of oxygen — physiologically — and could not replace it. In-
cidentally, it i,s of interest to note that this is the only means I have yet
found to produce encystment in this form.
From this collateral experiment we are able to say that the low
J
1
I
I
8
1
•ta
HOURS/
HOUKS
LIMI/VATED
\
t
8
7
FK;. 3. The relation between initial pH and rates of division in sealed cultures.
division rates in darkness, as represented in Fig. 3. were probably due
to the decreasing oxygen tension. It is doubtful if tin's effect is cor-
related with H'-ion concentration.
In the light, on the other hand, a pronounced optimum at about pH
6.7 was apparent in all series. (This optimum has been, in another
experiment, definitely established as below pH 6.9 and above pi I 6.2.)
The difference in rate between growth in light and in darkness must
SIGNIFICANCE OF pH IN F.UGLEXA 175
he ilue either directly to the products of photosynthesis or to the in-
creased oxygen supply available from the decomposition of CCX. Since
this difference is much less marked in aerated cultures, as will he shown
later, it is obvious that most of the increased growth in the sealed cul-
tures is only indirectly due to photosynthesis. It is actually due to the
oxidation of foods in the medium, but is made possible by the oxygen
produced during photosynthesis. ILuylcna, combining as it does two
forms of nutrition, uses more oxygen in its normal metabolism than does
a completely holophytic organism — the latter always producing, in good
light, more oxygen than is required in its own metabolism. If this
latter statement were not true, of course, life on earth would be
impossible.
Reproduction is fairly rapid between pH 7.6 and 4.0, but there is a
pronounced decrease in rate on both sides of the peak, — the rate at
the peak being as great as in initially aerated cultures. Photosynthesis
is, obviously, going on wherever growth in the light exceeds that in the
dark, and this difference in rate is proportional to the amount of photo-
synthesis. Therefore, it is apparent that, under conditions stated, a
definite optimum for photosynthesis exists in this form, and that it is
near pH 6.7. Furthermore, the oxygen production at this pH is great
enough to maintain reproduction at near the maximum rate observed in
media in equilibrium with the air.
While making the counts at the end of the experiments just described,
I observed in the most acid cultures a considerable number of organisms
attached to each other in pairs. The point of attachment was the
posterior tip, the last part to divide in longitudinal fission. I repeated
this observation with a stock culture, of pH 2.9, heavily inoculated with
Euglcna, and maintained under constant illumination. This was first
examined after forty-eight hours. At this time many of the organisms
present were attached in pairs, as previously described, but. in addition,
there were groups of three and four individuals composing these mul-
tiple monsters — and always all individuals were joined at the same point,
the posterior tip. Table II gives their relative numbers in samples of
uniform volume counted at the end of forty-eight hours after inocula-
tion. Figure 4 consists of camera lucida drawings of several of these,
and one or two other abnormalities which appeared with them. In
every case in which the individuals were of approximately equal size,
the connection was purely at the surface — no cytoplasmic connection
being observable even with most careful examination under oil im-
mersion.
The " colonies " of three or more individuals assumed the shape of
rosettes, in appearance reminding one of small colonies of colonial alg?e.
176
GORDON ALEXANDER
TABLE II
Relative Numbers of Single Individuals and Multiple Monsters in Ten Equal Volumes,
from a Culture at pH 2.9 Examined 48 Hours after Inoculation
Multiple Monsters
**M n cl t* In<_li viii uiil s
Double
Triple
Quadruple
24
11
1
25
13
16
12
1
21
9
1
23
9
19
8
4
28
17
1
1
23
18
3
2
23
7
1
23
14
Average 22.5
11.8
1.1
0.4
Their progress through the medium was very irregular, depending ap-
parently on the resultant of the vectors represented in the aggregate.
Groups of as many as six individuals were ohserved, always attached
at the point last to divide in their typical form of reproduction.
The effect is entirely on the surface, and is dependent on the H+-ion
concentration, — since it always occurs to a greater or lesser extent below
a pH of about 3.5, and is equally common in illuminated or darkened,
sealed or aerated cultures. It is apparently permanent. Subsequent
FIG. 4. Camera lueida drawings of monsters associated with media of low
].H. 1-3, typical double monsters; 4, a triple monster; 5-6, other types of abnor-
malities observed. In 5, the cytoplasm was continuous between the "bud" ""A
the parent organism.
and
SIGNIFICANCE OF PH IN EUGLENA 177
examinations of the stock culture showed that most, if not all, of these
multiple monsters never separated into individuals, but sank to the
bottom and died in a few days. The cultures slowly developed, how-
ever, presumably habilitated by those individuals that had not been
affected.
As the culture began to assume again the appearance of life (there
was a period of a few days in which, to the eye, it seemed to have died
out) an additional morphological effect of low pH began to be evident.
The organisms now developing in the culture could be distinguished, but
were not green. Examination under the microscope showed apparently
complete etiolation, commonly extending in part to the stigma. The
stigma was much reduced in size, — indistinguishable, in fact, in some
individuals. Although several causes of etiolation have been found in
Euglena gracilis, that due to low pH has not been previously reported.
Zumstein brought about etiolation by keeping cultures in an organic
medium in complete darkness. I have repeated this, using sterile pure-
line cultures. Zumstein also stated that the green form became colorless
in "' very rich organic medium." Ternetz described another hyaline
form, differing from those observed by Zumstein in being permanently
colorless. This form had completely lost the stigma, as well as chloro-
phyl. The type of etiolation associated with low pH is, however, not
permanent. It may be similar to that caused by darkness or a " very
rich organic medium," or may be identical with the latter, or with both.
In any case, inoculation of culture medium at pH 7.2 with etiolated
organisms from that at pH 2.9 yielded a normally green culture in a
little more than a week.
At about this same time a stock culture was prepared to use in a
study of the possible effect of initial adaptation in changing the shape
of the pH — division-rate curve. This culture was adjusted to pH 4.6.
As in the more acid cultures etiolation took place. When used for
inoculation, the culture did not appear green to the eye, but when held
to the light showed a rich growth of organisms. Under the microscope
these individuals appeared colorless.
I have no explanation to offer as to the mechanism of this effect,
but it is certainly associated with a high concentration of H+-ions.
Etiolation of Euglena gracilis ordinarily occurs in darkness at any pH,
but, even in the light, and with continuous illumination, it occurs below
pH 5 ±. This etiolation, either in darkness or at low pH in the light,
is evident on microscopical examination after only forty-eight hours.
ITS
GORDON ALHXAXUMR
Aerated Cultures
The next experiments were designed to provide a sufficient supply
of oxygen in both light and darkness. Figures 5 to 7. inclusive, repre-
-< -nt typical results, — the two former under conditions of initial aeration
only, and Fig. 7, under conditions of continuous aeration as compared
with initial aeration only, in identical cultures inoculated at the same
time.
Air from a storage tank was bubbled through the cultures at a
1
8
si
/
e
9
FIG. 5. The relation between initial pi I and rates of division of normal
ii'reen Eiit/lcna, in cultures initially aerated, and exposed to the air through cotton
plugs.
vigorous rate, with shaking, for two minutes. The cultures were in
flasks, each containing 80 cc. of medium and each adjusted to a par-
ticular pi I. The aeration was carried out immediately after the intro-
duction of organisms for inoculation. Following aeration, most of each
culture was divided among three or four sterile test-tubes plugged with
cotton, each containing 15 to 20 cc., — two of these to be illuminated, and
the other or others for a control in darkness. ( During the experiments
these cultures had access to the air through cotton plugs.) The pi I
was then determined in the medium remaining, and the final portion
was used for determining the density of organisms. Fach culture was
shaken just before each transfer in order to maintain uniform distribu-
SIGNIFICANCE OF P!l IN FUGLKNA
170
lion of the contained organisms. Agreement in counts from culture to
culture was very close, and. since the initial volume of inoculating cul-
ture was the same in each, this indicates that the samples counted fairly
represented the cultures used in the experiments.
The results from the initially aerated cultures are very different,
both under illumination and in darkness, from those obtained from
sealed cultures. In the experiment plotted in Kig. 5. in which normally
green organisms were used, there was, above pi I 4, a surprisingly small
difference in division-rate between those in light and those in darkness.
This, of course, is conclusive evidence that the major growth of this
liucilcna in the medium used is dependent on saprophytism, and not
FIG. 6. The relation between initial pH and rate of division of etiolated
Euylcna, in cultures initially aerated, and exposed to the air through cotton plugs.
photosynthesis. However, there is a slightly greater growth in the light
between pH 8.1 and 4.6. This is less than 10 per cent greater; but in
another experiment I found a difference of 15 per cent in the same
direction, at pll 7.5 and 6.6, under conditions otherwise the same. It
is probably, therefore, a real difference, and indicates that photosynthesis
plays some part in growth.
At about pll 4. on the other hand, the curve of growth in the light
falls away rapidly, whereas that of cultures in the dark remains at a
fairly high level. This indicates the presence of a photodynamic com-
pound.
Comparison of Fig. 5 with Fig. 6 gives a clue to the nature of this
compound. In Fig. 6. the results of experiments begun with etiolated
180
GORDON ALEXANDER
individuals are recorded. At the alkaline end of the range there is,
again, evidence of photosynthesis. This is not unreasonable, for ex-
amination ft" the Euglcna under the microscope showed the presence
of .small quantities of chlorophyl, even after only forty-eight hours in
the cultures. On the other hand, there is no evidence of photosynthesis
1>el"\v pH 6. There is some evidence of a photodynamic effect toward
the arid end of the range, but it is not nearly so pronounced as in the
Cewr/, wot/s
A£K 4 T/OM
9 / &
/NJT/AL f)M
FIG. 7. The relation between initial pi I and rates of division in cultures
continuously aerated e. .mpared with that in cultures only initially aerated but
subsequently exposed to the air through cotton pities.
The flat to-,) of the cur\< -dilative of continuously aerated cultures is
based mi another experiment, in which maximum rates were observed also between
pi I 7 and <S, and between pi I 5 and 4. For convenience, only four pH values were
selected in each " run." The very low rate of the initially aerated, illuminated
culture at pH 8.1 + was due to bacterial contamination.
ri< 3 li'-.^un with green organisms. This suggests that the photosensi-
tizing material is either chlorophyl or one of its derivatives. Since
etiolation occurs at about the- same pi I values, we cannot be certain that
the chlorophyl itself is the material. In any case, it seems strange that
an organism could be made photosensitive by one of its normal con-
st it uvnts. As far as I know, this is the first report of this effect. A
phot.idynamic effect of chlorophyl (production of hemolysis of mam-
malian erythrocytes) has been demonstrated by llausmann (1909), but
SIGNIFICANCE OF PH IN EUGLENA 181
a comparison with the observation herein recorded seems rather far-
fetched.
The stock culture used to inoculate the series plotted in Fig. 5 was
at pH 7.9, that used in the experiment with colorless Euylena, at pH 4.7.
In spite of this initial difference in pi I, no very evident adaptive effect
of the low pH in the stock culture is apparent. The curves are not
strikingly different except at the extreme acid end of the range, and
there the principle difference occurs only in the light.
As in sealed cultures, the undivided, multiple organisms appeared
in the more acid conditions.3 In one of these initially aerated cultures
the group of six, mentioned previously, was observed. This group can
have evolved as follows : The first individual divided partly, forming
individuals a and b. Then each of these divided in the same incomplete
fashion, forming a. a', b and b'. Probably two of the four now attached
divided, forming the six. There are alternative explanations, but this
is the more probable one, since it means a maximum of three divisions
each for four of the final six individuals, with two divisions each for
the other two. The number of divisions is in agreement with the
average for the culture.
In order to determine the validity of conclusions based on cultures
aerated only at the beginning of experiments, several series of tests were
carried out with cultures through which air was bubbled continuously.
For convenience, only four H+-ion concentrations were selected in each
experiment. As controls for the continuously aerated cultures (in both
light and darkness) duplicates were initially aerated and closed with
cotton plugs.
Figure 7 represents, the results of such a test. The continuous
aeration does produce a somewhat greater rate of division, due to the
large supply of carbon dioxide and oxygen, but the curves representing
the rates under such conditions are similar in shape to those obtained
from cultures only initially aerated. We may conclude from this that
the effects of pH in cultures initially aerated and having access to the
air are not qualitatively different from those in cultures continuously in
equilibrium with the air. The results here are, however, based on ex-
periments lasting for only forty-eight hours. Over longer periods of
time, the changes in standing cultures might tend to make the pH-
division-rate curve approach the form of that found for sealed cultures.
Observations on aerated cultures furnish little or no evidence for a
pronounced optimum for the rate of division. This rate is fairly uni-
3 All multiple organisms were counted as if all divisions begun had been
completed. Inasmuch as these divisions are probably never completed, but fol-
lowed by death, this method eliminates part of the true depression of division-rate
at low pH.
13
182 GORDON' ALEXANDER
form over an extremely wide range of H*-ion concentrations. A slight
increase between pH 6 and 7 suggests an optimum there, but this is not
nearly so apparent as that associated with photosynthesis alone.
All types of experiments consistently give us three explanations for
the depression of division rate at low pH, exclusive of that associated
with fatal injury to the organisms. (1) Etiolation. (2) production of
multiple monsters which subsequently die, and (3) a photodynamic
effect, all probably contribute to this depression. In addition, both high
and low pH values are fatal — the limits found being pH 2.3 and 1 1.0 -(-•
(Eitglena gracilis will grow at pH 9.0. but its average rate of division
is less than once in three days. Xo evidence of special effects at high
pH has been obtained.)
It may be thought that the form of movement commonlv observed
in Euglcna, and called variously " euglenoid movement" or 'Meta-
bolic," may be induced bv particular concentrations of hydrogen ions.
In 1925, while working with Euglcna gracilis, I found that organisms
transferred from an acid medium t<> one distinctly more alkaline, or
rice versa, began to go through this type of rhythmical contraction, and
did not return to the normal swimming form for some time. The re-
sponse is quite evidently associated with a change of environment, rather
than with the nature of the new environment. It seems to the writer
that the respmisi- to pH indicates that "euglenoid movement' is a
modified avoiding reaction. It is certainly not a form of locomotion,
for as a mean-- of progre» I nun place to place it is little, if at all. more
efficient than Brownian movement.
Sl'M MARY
1. The alleged tolerance of high concentrations of citric acid by
I'.itglcnu (/racilis Klebs is associated with Il'-ion concentration. This
species of Eitglena tolerates Hf-ion concentrations as high (or higher)
when produced by hydrochloric, sulphuric or oxalic acids.
2. Even less acid solutions of valeric and salicylic acids cause death,
however. This effect may be connected with the greater rates of pene-
tration of these acids into living cells.
3. Increased growth of I-'nglcua (/racilis at high acidities is due to
the elimination of bacterial or other competition at such pH values. It
does not occur in a series of sterile cultures.
4. A sterile, pure-line, well-aerated culture of Euglcna gracilis, in
Bacto-peptone medium, showed little difference in division-rate between
pll 7.7 and 4.5 in the light, and between pi I 7.7 and 3.0 in darkness.
An only slightly greater rate of division at about pll 6.7 suggests that a
true optimum may exist at that point.
SIGNIFICANCE OF pH IN EUGLENA 183
The absolute limits of life proved to be approximately pH 2.3 and
11.0 -f.
5. There is a marked optimum pi I for photosynthesis in this
Euglena. Sealed cultures in the light invariably developed maximum
growth at pH 6.5 to 6.8. Comparison with aerated cultures in darkness
shows that the difference between growth of sealed cultures in light and
in darkness is actually not due to the products of photosynthesis in the
former, but to the use of the oxygen produced during photosynthesis in
the oxidation of organic foods in the medium.
6. Quantitatively, enough oxygen is produced at pH 6.7, in photo-
synthesis by this Euglena to maintain its own reproduction at maximum
or nearly maximum rate. Euglena may, therefore, be a significant
element of the carbon cycle where it occurs in nature.
Under conditions of nearly complete anaerobosis in darkness it does
not grow, but encysts, and subsequently dies.
7. In cultures of about pH 3.5 or lower, many individuals fail to
complete normal division, remaining attached together by the surface
at the posterior tip (the last part to divide in euglenoid fission). Two.
three, four and even as many as six individuals have been found at-
tached together, the higher numbers of individuals forming rosettes.
This condition depends on pH, as such, occurring equally in light and
darkness, in sealed and aerated cultures. Furthermore, it is apparently
only a surface effect, as, although occasionally other types of abnor-
malities appear, in the type here described there is no connection between
the cytoplasm of one individual and that of another.
8. More or less complete etiolation occurs, even under continuous
illumination, below pH 5 ±. This is not permanent, the chlorophyl
reappearing after the organisms are returned to a medium of lower
H+-ion concentration.
9. In cultures of low pH (below about pH 3.5) a photodynamic
effect is present. This is far more marked with green than with
etiolated individuals, which suggests that the contained chlorophyl, or a
derivative of it, is the photodynamic compound involved.
10. Encystment does not occur as an effect of pH anywhere in the
range studied, pH 8.5 to 2.4.
11. The so-called "euglenoid movement'1 is not induced by any
particular pH, but may be brought about by transferring the organisms
from the culture in which they have been living to one in which the pH
is markedly different. This type of movement should not be considered
a form of locomotion, but, rather, a modified avoiding reaction.
184 GORDON ALEXANDER
BIBLIOGRAPHY
ALLEE, \Y. C., 1930. Concerning Community Studies. Ecology. 11: 621.
BLACKMAX, F. F., 1905. Optima and Limiting Factors. Ann. of Dot., 19: 281.
BRESSLAU, E., 1926. Die Bedeutung der Wasserstoffionenkonzentration fur die
Hydrobiologie. Vcrh. d. Int. Vcrcin. f. thcon-t. K. angcu'iindtc Limnologic.,
3: 56.
CHAI.KLEY, H. W., 1930. On the Relation between the Resistance to Heat and the
Mechanism of Death in Paramecium. Physiol. Zoo!., 3: 425.
DARBY, H. H., 1929. The Effect of the Hydrogen Ion Concentration on the
Sequence of Protozoan Forms. Arch. f. Protist., 65: 1.
DARBY, H. H., 1930. The Experimental Production of Life Cycles in Ciliates.
Jour. Ex per. BioL, 7: 132.
HARGITT, G. T., AND W. W. FRAY, 1917. The Growth of Paramecium in Pure
Cultures of Bacteria. Jour. E.vpcr. Zoo!., 22: 421.
HARVEY, E. N., 1928. Photosynthesis in Absence of Oxygen. Plant Phvsiol., 3:
85.
HAUSMANN, W., 1909. Die photodynamische Wirkung des Chlorophylls und ihre
Bezieliung zur photosynthetischen Assimilation des Pflanzen. Jahrb. f.
n'iss. Bot., 46: 599.
JAHN, T. L., 1929. Studies on the Physiology of the Euglenoid Flagellates.
I. The Relation of the Density of Population to the Growth Rate of
Euglcna. Bio!. Bull., 57: 81.
KLEBS, G., 1883. t;ber die Organisation einiger Flagellatengruppen und ihre
Beziehungen zu Algcn und Infusorien. Untcrsuch. aus d. hot. Inst. zu
Tiibingcn., 1 : 233.
KOFFMANN, M., 1924. tjbcr die Bedeutung dcr \Yasscrstoffionenkonzeritration
fiir die Encystierung bci einigen Ciliatcnarten. Arch. f. mikr. Atwt.,
103: 168.
KOST'R, W. J., 1921. The Comparative RfMMancc of Different Species of
Euglenidae to Citric Acid. Ohio Jour. Science, 21: 267.
LOOM is. A. L., E. N. HARVEY AND C. MACRAE, 1930. The Intrinsic Rhythm of
the Turtle's Heart Studied with a New Type of Chronograph, Together
with the Effects of Some Drugs and Hormones. Jour. Gen. Phvsiol.,
14: 105.
PARPART, A. K., 1928. The Bacteriological Sterilization of Paramecium. BioL
Bull., 55: 113.
PHILPOTT, C. H., 1928. Growth of Paramecia in Pure Cultures of Pathogenic
Bacteria and in the Presence of Soluble Products of Such Bacteria. Jour.
Morph. and Physiol.. 46: 85.
POWERS, E. B., 1930. The Relation Between pH and Aquatic Animals. Am. Nat.,
64: 342.
ROBERTSON, T. B., 1922. Reproduction in Cell-Communities. Jour, of Phvsiol.,
56: 404.
SKADOWSKY, S. N., 1926. tvber die aktuelle Rcaktion der Susswasserbecken und
Hire biologischc Bedeutung. Vcrh. d. Int. Vcrcin. f. thcorct. u. angewandte
Lininolocjic, 3: 109.
SPOEHR, H. A., 1926. Photosynthesis. New York.
STILES, WALTER, 1925. Photosynthesis. The Assimilation of Carbon by Green
Plants. London.
TERXETZ, 'CHARLOTTE, 1912. Beitriige zur Morphologic und Physiologic dcr
Euglcna gracilis Klebs. Jahrb. f. u'iss. Bot., 51: 435.
WAXN, F. B., AND E. F. HOPKINS, 1927. Further Studies on Growth of Chlorella
as Affected by Hydrogen-ion Concentration. Bot. Gaz., 83: 194.
WARBURG, OTTO, 1919. Cher die Geschwindigkeit der photochemischen Kohlcn-
saurczersetzung in lebendcn Zcllcn. Biochem. Zcitschr.. 100: 230.
ZUM STEIN. HANS, 1899. Zur Morphologic und Physiologic der Euglena gracilis
Klebs. Jahrb. f. ii'iss. Bot., 34: 149.
THE RESPIRATORY FUNCTION OF THE BLOOD OF
(JRECHIS CAUPO
ALFRED C. REDFIELD AND MARCEL FLORKIN 1
(From the Hopkins Marine Station, Pacific Grorc, California)
Ureclus caupo is an echiuroid worm inhabiting sandy mud flats in
the estuaries of the coast of California. Its characters and habits have
been described by Fisher and MacGinitie (1928). Because of its large
size, the simple nature of its circulatory and respiratory systems, and
the fact that its ccelomic fluid is voluminous and contains abundant red
blood corpuscles, it affords unusually suitable material for the study of
respiratory problems. The present paper contains a description of those
properties of the blood of Urcchis which are of importance in respira-
tion, together with certain observations designed to evaluate the sig-
nificance of these properties.
The material used for these studies was collected in the Elkhorn
Slough, a tributary of Monterey Bay. The authors wish to express
their indebtedness to Dr. MacGinitie for assistance in procuring the
animals and to Professor Fisher for the many courtesies received while
they were at the Hopkins Marine Station.
I. THE BLOOD OF URECHIS
Fisher and MacGinitie (1928) state that the ccelom is filled with
bright red blood, the pigment being lodged in subcircular cells, about
0.025 mm. in diameter, which readily distort when crowded. There are
also very numerous amoeboid cells which when aggregated are yellow in
color. We have found the color of the blood to vary, being frequently
of a dull brownish-red color; less often, and particularly in smaller
specimens, of a bright scarlet resembling the blood of vertebrates. A
volume of 15 or 20 cc. may be secured from a single specimen.
The Blood
The plasma does not clot, and when separated from the cells is a pale
yellow color ; not infrequently it may be tinged with the corpuscular
pigment. Under microscopic examination the cytoplasm of the cor-
puscles appears yellow and is seen to be filled with small, highly refrac-
tive granules. In addition, there are many granules of a brown pigment
1 Fellow of the C. R. B. Educational Foundation.
185
186
A. C. RRDFJKI.D AXD M. FLORKIX
in the corpuscles of some specimens. The occurrence of this pigment
is variahle and it may be nearly lacking in those specimens whose blood
appears scarlet rather than brownish-red. In the center of the corpuscle
is a clearer area, which, on staining, proves to be a small nucleus. The
cu-lomic cells of Urcchis chelensis described by Seitz (1907) are ap-
parently similar, being nucleated and containing yellow pigment granules
( in preserved material).
TAHLE I
Spectrometric Data, of Urechis Hemoglobin
Wave
Length
Oxyliemo-
globin
Reduced
Hemoglobin
Wave
Length
Oxyhemo-
globin
!<• luced
Hemoglobin
mn
E
E
mji
E
E
450.4
0.394
0.495
566.6
0.221
0.299
460.5
0.280
0.225
567.6
0.223
470.6
0.226
0.169
569.7
0.245
0.273
480.7
0.189
0.160
571.7
0.2S5
490.8
0.179
0.164
573.7
0.297
500.9
0.174
0.176
575.7
0.309
511.0
0.172
0.197
576.7
—
0.238
516.1
0.180
577.8
0.310
521.1
0.192
0.218
578.9
0.297
523.2
0.206
581.8
0.267
0.204
525.2
0.215
583.8
0.214
527.2
0.231
585.8
0.168
•
529.3
0.244
586.9
—
0.171
531.3
0.264
0.245
587.9
0.131
533.3
0.276
589.9
0.102
535.3
0.291
591.9
0.082
0.145
537.3
0.299
597.0
0.055
539.3
0.306
602.0
0.042
0.096
541.4
0.312
0.277
612.1
0.032
0.073
543.4
0.310
622.2
0.025
0.064
545.4
0.292
632.4
0.022
0.059
546.4
—
0.298
ol2.5
0.024
0.056
547.4
0.276
652.6
0.009
0.049
519.5
0.255
662.7
0.016
0.046
551.5
0.243
0.313
672.8
0.004
0.038
555.5
0.214
682.9
0.008
0.036
556.5
—
0.313
693.0
0.003
0.028
559.6
0.197
561.6
0.197
0.312
563.6
0.202
. 565.6
0.211
The corpuscles appear to be surrounded by a strong membrane. On
dilution of blood with distilled water, the cells swell but do not burst.
Upon applying pressure to tlu- cnverslip when in this condition the
membrane ruptures and the contents may be seen to flow out through
a localized opening. The granules in the swollen corpuscles are in active
Brownian movement, suggesting a fluid state of the interior of the cell.
RESPIRATORY FUNCTION Ol- URECHIS I'.LOOD
187
In a one per cent sapunin solution the cells swell, the granules remaining
confined to the previous volume of the cells and appearing surrounded
by a clear region. After a few minutes the membrane spontaneously
ruptures and the granules flow out through the localized opening.
450 500 550 600 650 700 mM.
Wave Length
FIG. 1
450 500 550 600 650 700m,/.
Wave Length
FIG. 2
FIG. 1. Absorption spectrum of oxygenated hemoglobin of Urcchis caupo.
Ordinates : extinction coefficient of solution of unknown concentration.
FIG. 2. Absorption spectrum of reduced hemoglobin of Urcchis caupo.
Ordinates : extinction coefficient of solution of same concentration as that shown in
Fig. 1.
The Respiratory Pigment
That the corpuscles of UreeJiis contain hemoglobin is indicated by
the spectroscopic examination of laked blood. Typical hemoglobin
crystals may be obtained by allowing laked corpuscles to dry under a
coverglass. For spectrophotometric examination a solution of hemo-
globin was prepared by laking one cc. of corpuscles separated by centri-
fugation with 11 cc. of distilled water containing three drops of ether.
To this solution was added 8 cc. of 4M ammonium sulfate brought to
ca. pH 8 by the addition of ammonia. From this solution the corpus-
cular debris was filtered off and one volume of the filtrate diluted with
four volumes of water. Filtration was repeated. The filtrate so ob-
tained contained the original corpuscular content of hemoglobin diluted
1 : 100 and in the presence of 0.8 molar ammonium sulfate at a pH of
approximately 8.
1SS
A. C. RKDFIELD AXD M. FLORKIX
One sample of this solution was employed for measuring the absorp-
tion spectrum of oxyhemoglobin ; another specimen was reduced by
equilibration with hydrogen and used to obtain the spectrum of reduced
hemoglobin. The solutions were perfectly clear. The measurements
were made with a Konig-Martens spectrophotometer within five hours
of the completion of the preparations. The length of the column of
fluid was 3.3 cm. The extinction coefficients of these solutions, esti-
mated for one cm. length, are recorded in Table I and illustrated
graphically in Figs. 1 and 2. In Table II are recorded the wave length
of maximum density in the a and ft bands and the wave length of mini-
mum absorption between these bands as obtained by various workers
with various hemoglobins.
TABLE II
Spectrometric Characteristics of Various Oxyhemogldbins
Species
Wave Length of
Maximum
Absorption
Wave Lrnnth of Min-
imal Absorption be-
Ratio of Extinction
Coefficients at
Maximum 0 and
Observer
t\vt?en cc and Q Band1^
Alinimum between
a and 0
a Band
ft Band
m/i
WM
>MH
Dog and
horse. . . .
575.6
540.4
558. 1
>1.60
Hari (1917)
Dog
575 577
539-542
560.*
1.63
Kennedy
(1926-27)
Horse
578.2
540.4
562.5*
1.58*
Vies (1923)
. \rcnicola . .
576.0
540.0
560.*
1.53*
Vies (1923)
Marphysa .
57S.O
54o.'i
Vies (1923)
Cucumaria
57').
542.7
55S.
Van der I.ingen
and Hoghen
(1928)
t'rechis. . . .
577.
542.
561.
1.58
* Estimated from published data of observer.
These values for Urcchis hemoglobin agree more closely on the whole
with those found for the hemoglobin of the worm Marphysa and the
holothurian Cucumaria and with Vies' measurements of horse hemo-
globin than with this author's data for .Ircnicola. The values for horse
hemoglobin obtained by Vies, from \\hich lie concluded that Arenicola
hemoglobin differed from horse hemoglobin, do not agree with the values
obtained by Hari and Foster for the mammalian pigment, which are very
similar to Vies' values for Arenicola. The shape of the absorption
curve for Urcchis does not agree exactly with the data recorded for
mammalian oxyhemoglobin, particularly in the region about 510 m/u.
Discrepancies in the shape of the curves may be attributed to the
RESPIRATORY FUNCTION OF URECHIS BLOOD
189
presence of methemoglobin in the solutions, as Ilari has pointed out; a
fact which makes the direct comparison of the curves difficult. The
data leave no doubt that the pigment of the Urechis blood is a hemo-
globin, but the spectrometric evidence regarding the specificity of the
hemoglobin cannot safely be interpreted.
TAULK III
Cell Volume and Oxygen Capacity of Urechis Blood
Specimen
Volume Red Cells
Oxygen Content
Oxygen Combine 1
Oxygen Combined
per 100 cc. Cells
per cent
volumes per cent
volumes per cent
cc.
3
36.6
—
—
—
4
26.4
—
—
—
7
40.3
6.30
5.80
14.4
7.22
6.72
16.7
8
37.6
5.83
5.33
14.2
5.70
5.20
13.8
5.77
5.22
13.9
9
18.3
2.S5
2.35
12.8
2.87
2.37
12.9
10
35.3
5.72
5.22
14.8
5.53
5.03
14.2
12
23.8
4.64
4.14
17.4
4.36
3.86
16.2
4.24
3.74
15.7
13
19.5
3.89
3.39
17.4
3.70
3.20
16.4
14
23.2
2.90
2.40
10.3
2.83
2.33
10.0
2.66
2.16
9.3
15
28.6
4.53
4.03
14.1
4.30
3.80
13.3
16
32.0
4.54
4.04
12.6
4.78
4.28
13.4
21
—
3.70
3.20
—
20
4.43
3.93
—
4.50
4.00
—
23
4.09
3.59
—
—
4.05
3.55
—
A. c. RI-:DFIELD AND M. FLORKIN
Hemoglobin appears to occur in the musculature of L'rccliis, par-
ticularly in that of the foregut, or crop. In this structure, which is in
a thin muscular tube of a bright pink color, the spectrum of oxyhemo-
globin can be beautifully demonstrated with the microspectroscope. If
the preparation is covered with a coverglass, the spectrum soon changes
to that of reduced hemoglobin, except near the edges, where the oxy-
hemoglobin bands persist. Because of the absence of capillaries in this
preparation it should form very advantageous material for the study of
the function and properties of muscle hemoglobin.
The Quantity of Corpuscles and Hemoglobin in flic Blood
The red corpuscles occupy from 18 to 40 per cent of the total volume
of the blood when separated with the hematocrit (Table III). A gray
layer of rather variable volume containing sperm or eggs and other cells
separates between the red cells and the plasma. The oxygen content of
the blood equilibrated with air was determined with the Van Slyke con-
stant volume blood gas apparatus, using one cc. samples, and varies
between two and six volumes per cent. Special care was taken to stir
the blood before sampling because of the rapid rate at which the large
corpuscles settle out. A one per cent saponin solution was used as lak-
ing reagent. These values are recorded in Table 111 and may be com-
pared with the values found for other worms and other invertebrates
containing hemoglobin in Table IV.
TABLE IV
Oxygen Content of Blood of Worms and Other Invertebrates
(equilibrated with air)
Species
< >\\Ken
< <>iu'-m
1' nit-Ill
1 i. . in rence
Observer
L rechis caupo
2.66-7.22
1 lemoglobin
in corpuscles
( ilycera siphonostoma
Arenicola sp
2.58 3.03
5.70 8.70
1 lemoglobin
I lemoglobin
in corpuscles
in solut ion
Winterstein (1909)
Fox (1926) after Bar-
Cardita sulcata
1 2
1 lemoglobin
in solution
croft and Bancroft
(1924)
Winterstein (1909)
Pectunculus violaceus
Spirographis
1-2
8 Mi 10.0
1 lemoglobin
("hlorocruorin
in solution
in solution
Winterstein (1909)
Fox (1926)
Siphunculus nudus. . .
ca. 2
1 Irmeryt hrin
in corpuscles
\\mterstein (1909)
It is commonly believed that the inclusion of the respiratory pig-
ments within corpuscles has made possible the superior oxygen capacity
of the blood of vertebrates. This possibility does not appear to have
been realized in the invertebrate stage of development, for Arenicola and
RESPIRATORY l-VMTION OF URKCHIS BLOOD
Spirographis, which carry their respiratory pigments in solution, have a
greater oxygen content than Urechis and the other invertebrate forms
in which oxygen is transported in blood corpuscles.
The concentration of hemoglobin in the corpuscles of Urechis ap-
pears to be much less than is the case in vertebrates. In Table III is
recorded the estimated oxygen combined per 100 cc. of red corpuscles
—allowance being made for 0.50 volumes per cent of oxygen assumed to
l)e present in solution. The oxygen-combining power of the cells varies
from about ten to seventeen volumes per cent. Drastich (1928) finds
the following values for the hemoglobin content of the cells of verte-
brates: — various mammals 29.5 to 34; various birds 29.54; Rana
csciilcnta 24.85 ; carp 26.02 grams per 100 cc. corpuscles. Assuming the
Urechis hemoglobin to have the same oxygen-combining power per unit
weight as mammalian hemoglobin (one volume per cent oxygen capacity
corresponding to 0.746 grams of hemoglobin per 100 cc.), Urechis cor-
puscles are estimated to contain 7.5 to 12.7 grams of hemoglobin per
100 cc. of cells. The Urechis corpuscle is then about one-third as ef-
fective in transporting oxygen as those of the vertebrates. It is to this
fact rather than to a deficiency in the number of corpuscles that the low
oxygen capacities of the blood are principally due.
TABLE V
Data on Equilibrium of Oxygen with Urechis Blood
Carbon Dioxide
Pressure
Oxygen
Pressure
Oxygen
Content
Oxygen
Dissolved
Oxygen
Combined
Saturation
mm. Hg
mm. Hg
volume per cent
volume per cent
volume per cent
per cent
8.64
5.98
0.63
0.192
0.44
19.1
9.78
8.50
0.87
0.027
0.84
36.6
8.61
12.15
1.15
0.039
1.11
48.3
7.90
16.35
1.46
0.053
1.41
61.3
7.54
23.08
1.88
0.074
1.81
78.8
10.87
41.85
2.09
0.135
1.95
84.8
6.54
48.15
2.15
0.155
1.99
86.6
6.11
72.61
2.32
0.234
2.09
90.9
7.56
87.21
2.36
0.281
2.08
90.4
air
air
2.90
0.50
2.40
2.83
0.50
2.33
2.66
0.50
2.16
av. 2.30
100
Iron Content of Blood
Attempts to estimate the iron content of the blood by the method of
Hall and Gray ( 1929) yielded rather discordant results. The values
obtained were always of the order expected from the oxygen capacity
of the samples.
192
A. C. REDFIELD AND M. FLORKIN
The Equilibrium of Oxygen icith the Blood
The oxygen dissociation curve of the whole blood has been deter-
mined u>ing the Van Slyke constant volume apparatus for blood gas
estimations and the Haldane analyzer for measuring the composition
of the gas with which the blood has been equilibrated. Equilibration
was carried out upon 3 cc. of blood enclosed in 250 cc. tonometers
rotated for 20 minutes in a water bath at 19° C. Analyses were made
•*
immediately after equilibration in fear that the metabolism of the cells
might alter the gaseous content were the samples allowed to stand.
The carbon dioxide pressure was maintained approximating that obtain-
ing in the blood in riro; about seven millimeters. The data are recorded
in Table A".
In estimating the combined oxygen from the oxygen capacity it is
assumed that blood in equilibrium with air dissolves 0.5 volumes per
cent oxygen, the solubility at lower oxygen pressures being proportional
in accordance with llenry's law. The oxygen dissociation curve is
plotted in Fig. 3.
100
80
60
40
20
c
:
r
0 10 20 ^ii 40 50 60 70 80 90
' Ky.ni-ii I'ressure
FIG. 3. Oxygen dissociation curve of blood of L'rcchis canf>o. Temperature
19° C. For data see Table V. Ordinates : percentage of saturation; abscissae:
partial pressure of oxygen in mm. Hg.
RESPIRATORY FUNCTION OF URECHIS BLOOD
193
The Effect of Carbon Dioxide upon tJic Oxygen Dissociation Curve
Samples of blood have been equilibrated with oxygen in the presence
of carbon dioxide at pressures varying from 0.54 to 92 nun. Hg. The
temperature of equilibration was 19° C. The results are recorded in
Table VI. From this data, p-(1, the oxygen pressure at which the
blood would have been half saturated with oxygen has been calculated,
assuming the curves to have the same shape as that drawn in Fig. 3.
TABLE VI
Data on Equilibrium of Blood with Oxygen at Various CO2 Pressrires
Specimens
Carbon
Dioxide
Pressure
Oxygen
Pressure
Oxygen
Content
Oxygen
Dissolved
Oxygen
Combined
Satura-
tion
pat
mm. II g
mm. HR
vol.
per cent
vol.
per cent
vol.
per cent
vol.
per cent
mm. Hg
0.54
10.40
1.96
0.03
1.93
49
10.6
0.94
14.89
2.62
0.04
2.58
66
10.7
8.60
7.10
2.16
0.02
2.14
55
6.3
Urechis No. 15
8.82
14.12
2.60
0.04
2.56
66
10.1
19.60
8.40
2.25
0.12
2.23
57
7.2
19.80
16.25
2.68
0.05
2.63
67
11.3
29.40
12.55
2.20
0.04
2.16
55
11.2
air
air
4.53
0.50
4.03
103.3
—
4.30
0.50
3.80
97.4
• — -
0.62
8.38
1.74
0.03
1.71
41.
9.9
0.71
10.80
1.95
0.03
1.92
46.
11.7
0.76
9.56
1.96
0.03
1.93
47.
10.1
Urechis No. 16
1.26
9.44
1.77
0.03
1.74
42.
11.1
77.0
9.08
2.14
0.03
2.11
51.
9.0
92.0
11.30
2.34
0.04
2.30
55.
10.1
air
air
4.54
0.50
4.04
97.
—
4.78
0.50
4.28
103.
—
These values, recorded in the last column of Table VI, make it
appear that the affinity of the blood for oxygen is not influenced to a
detectable degree by the pressure of carbon dioxide within the ranges
of pressure examined. In this regard the blood of Urechis differs from
that of most vertebrates and from that of Arenicola. In the latter form
Barcroft and Barcroft (1924) found the typical effect of hydrogen ion
concentration upon the oxygen dissociation curve. Recently Dill and
Edwards (1931) have observed that in the blood of the elasmobranch,
Raid oscillata, the effect of carbon dioxide upon the oxygen dissociation
curve is absent or nearly so.
104
A. C. REDFIELD AND M. FLORK1X
The Effect of Temperature upon the O.vyyen Dissociation Curve
Oxygen dissociation curve data obtained from the same specimen
of blood have been secured at two temperatures, 22° C. and 34.5° C.
i Fig. 4). The carbon dioxide tension was about 12 mm. in both cases.
10
20 30 40
( >xygen Pressure
60
I;i<;. 4. Oxy^rn dissociation curves of blond of I'rccliis cii'.ipn. r<|iiilibratcd
at temperatures of 22° C. and 34° C. Ordinates : percentage of .saturation;
abscissa?: partial pres.Mire of oxygen in nun. lit;.
The curve drawn through the data obtained at 22° C. is identical with
that in Fig. 2 obtained from another sample of blood at 19° C. At
34° C. the points lie well to the right. The data are insufficient to war-
rant any conclusion with regard to the shape of the curve at the higher
temperature, but it is clear that the temperature effect upon the oxygen
equilibrium is large and of the same direction and order observed in
vertebrate hemoglobin I Krown and Hill. 1923; Maccla and Seliskar,
1925).
The Equilibrium of Carbon Dioxide with the Blood
Table VII presents the data obtained by equilibrating Urcchis blood
against various mixtures of carbon dioxide in air at 18.5° C. The
analyses were made with the Van Slyke apparatus and the llaldane
analyser. The oxygen capacity of the blood employed corrected for
dissolved oxygen was 3.9 volumes per cent.
In order to facilitate comparison of the ['rechis blood with that of
RESPIRATORY FUNCTION OF URECHIS BLOOD
195
TAHLE VII
Data on Equilibrium of Carbon Dioxide with Urechis Blood
Carbon Dioxide
Pressure
Carbon Dioxide
Content
Carbon Dioxide
Dissolved
(HsCOs)
Carbon Dioxide
Combined
(BIICO3)
(BHCO.)
k (IfcCO,)
mm. Hg
vol. per cent
vol. per cent
vol. per cent
0.9
3.32
0.09
3.23
1.597
3.3
6.13
0.35
5.78
1.217
7.22
8.90
0.76
8.14
1.029
12.40
11.00
1.30
9.70
0.873
22.0
14.15
2.31
11.84
0.710
47.2
19.15
4.96
14.19
0.456
other animals, and for the comparison of various experiments with this
species it is convenient to relate the data to the logarithm of the ratio of
combined (BHCO3) to free (H2CO:{) carbonic acid. This function
changes approximately in proportion to the hydrogen ion concentration,
which may be obtained by adding the appropriate pK value. Moreover,
the total buffer value of the blood is also dependant upon this function.
The quantity of carbon dioxide dissolved in the blood or present as
H2CO3 (free carbonic acid) has been estimated assuming a, the number
of cubic centimeters of CO._, dissolved in one cubic centimeter at a
pressure of 760 mm. Hg, to be 0.80. This value is slightly less than
the value 0.827 given by Bohr (1897) for two per cent NaCl at 18° C.
The concentration of combined carbonic acid (BHCO3) is obtained
by subtracting the free carbonic acid (HoCO:!) from the total carbonic
acid. The estimated values of these quantities are included in the table.
The total buffer value of blood, /?, is denned by the equation
— A (BHCO,)
ft '' .. (BHCO.,)
AlogTHxaj
In Fig. 5 the values of (BHCO.,) are plotted against log
(BHCO3)
(HXO,)
Throughout a considerable range the points fall about a straight line,
indicating as in the case of mammalian blood that the buffer value is
constant. The value of ft is given by the slope of this line and is 11
volumes per cent (or 0.49 milliequivalents per liter).
It will be shown subsequently that the plasma of Urechis possesses
little or no buffer value. Is the total buffer value of the Urechis blood
adequately accounted for by the quantity of hemoglobin in the cor-
puscles? In this specimen of blood the oxygen capacity was 3.9 vol-
umes per cent. The buffer value per equivalent of hemoglobin is given
196
A. C. REDFIELD AND M. FLORKIX
by ft/3.9 == 2.82. This value is intermediate between the buffer values
of oxygenated and reduced hemoglobin as it occurs in the cells of the
blood of man and of the crocodile, the extreme values being 2.4 for
16
12
0.4
0.8
log [BHC03]/[H2C03]
1.2
FIG. 5. Relation of combined carbonic acid (BHCOO, to log
1.6
. inn o .
(II, CO,)
blood of L'l-ccliis canpo. Temperature 19° C. For data sec Table VII.
for
reduced crocodile blood at 29° and 3.47 for oxygenated blood of this
species (Dill and Edwards, 1931). The concentration of hemoglobin
appears sufficient to account for the total buffering effect of Urcchis
blood.
The effect of oxygenation and reduction upon the carbon dioxide-
combining power of blood is considered to be directly related to the
reciprocal effect of carbon dioxide (or hydrogen ion concentration) upon
the oxygen dissociation constant of hemoglobin (Henderson, 1928.
Chapter IV). We have shown in the case of Urcchis that the latter
RESPIRATORY FUNCTION OF URFCHIS BLOOD
197
is uninfluenced by the quantity of carbon dioxide present, and it is
consequently interesting to inquire whether reduced blood possesses the
same carbon dioxide-combining power as oxygenated blood. Dill and
Edwards (1931) state that both effects are absent or nearly so in the
blood of the skate, Raia oscillata.
As the effect is proportional to the concentration of hemoglobin, it
may be expected to be small in any case. In order to facilitate the ex-
periments, the blood of about ten animals was mixed and the corpuscles
separated with the centrifuge. A small quantity of plasma was mixed
with the corpuscles, yielding a solution containing 78 per cent red blood
corpuscles and having an oxygen capacity of 10.7 volumes per cent.
Were the Urcchis hemoglobin similar to mammalian hemoglobin the
reduced solution should combine four or five volumes per cent more
( P FT CO ^
carbon dioxide than the oxygenated serum when log ' „ - is 0.8.
\ "^ 2 ^~^ 3 /
Table YIII shows the result of equilibrating samples of this concen-
trated blood with air and with hydrogen containing about twenty-two
mm. pressure of carbon dioxide. The quantity of oxygen found in the
TABLE VIII
Data on Carbon Dioxide Equilibrium in Oxygenated and Reduced Blood
Oxygen Pressure
Carbon Dioxide Pressure
Carbon Dioxide Combined
mm. Hg
mm. Hg
vol. per cent
air
21.48
13.42
air
21.50
11.50
air
19.85
10.90
air
23.90
11.21
3.54
23.62
11.85
3.70
23.65
11.50
tonometers used for the " reduced " samples would not oxygenate more
than ten per cent of the hemoglobin. Disregarding the first experiment
of the series, the combined carbon dioxide is the same within the limits
of experimental error in both series.2 Certainly the phenomenon does
not occur with the magnitude commonly observed in the blood of the
higher vertebrates, and as in the case of the skate, it may be concluded
that the reciprocal effects of oxygen and carbon dioxide upon the
equilibrium of Urechis hemoglobin with these gases are absent or nearly
so.
- A second experiment in which the corpuscles were not so highly concentrated
and in which the measurements did not agree so closely yielded higher CO- values
for the oxygenated than for the reduced samples.
14
198
A. C. REDFIELD AXD M. FLORKIX
The distribution of carbon dioxide between the cells and plasma is
of particular interest as Urechis is the most primitive type of animal in
which the respiratory properties of the corpuscles have been studied.
The red blood corpuscle of Urechis is a much more " typical " cell than
are those of the vertebrates and one looks for properties which contrast
it with these more highly specialized erythrocytes.
TABLE IX
Data on Distribution of Carbon Dioxide between Corpuscles and Plasma
Specimen No.
19
22
24
25
26
Temperature of equilibration — ° C.
19.5
19.0
20
19
19
Volume of corpuscles — per cent ....
23.0
34.5
23
34.7
31.7
Low Pressure Experiment
COj pressure of equilibration — mm. Hg
8.0
12.95
10.50
10.20
9.31
Whole hlood-COj content — vol. per cent
8.82
14.20
11.40
11. IS
10.80
True plasnia-COj content — vol. per cent
7.09
14.15
11.30
10.95
11.45
Separated serum
COi pressure of equilibration — mm. //«
53.6
68.4
54.7
47.0
CO» content at this pressure — vol. per cent. . . .
//;;'/' Pressure Experiment
CO-> pressure of equilibration — mm H%.
11.74
62.5
19.30
62.2
16.54
49 5
15.40
46.7
Whole blood-COj content — vol. per cent
_';) 7
23 5
194
19.56
True plasma-CCK content — vol. per cent
18.9
20.4
17.12
18.82
Several experiments have been made in order to elucidate the re-
spective parts which corpuscles and plasma play in the transport of
carbon dioxide, and to determine the extent to which there is an ex-
change of material between cells and plasma. The procedure has been
to equilibrate blond with carbon dioxide at a pressure comparable to
that existing in vivo. With a part of this solution duplicate determina-
tions were made of the carbon dioxide content of the whole blood. The
remainder was centrifuged under oil in stoppered tubes. A portion of
the plasma so separated was analy/ed for the carbon dioxide content of
the true plasma, i.e., the plasma in equilibrium with corpuscles at the
original carbon dioxide tension. From these measurements together
with a measurement of the fraction of the whole blood occupied by the
corpuscles, made by hematocrit, the ratio of carbon dioxide concentra-
tion in corpuscles and plasma could be calculated. The remainder of
the plasma was then equilibrated with a relatively high CO2 tension in
order to give an idea of the buffer action of the separated plasma. The
foregoing measurements, which are designated as the "' low pressure
experiment " in the tables, were performed in the mornings on which
the blood was drawn. In the afternoon the "high pressure experi-
ment'1 was carried out. A portion of the whole blood was now
RESPIRATORY FUNCTION OF URFCHIS BLOOD
199
equilibrated at a relatively high CCX pressure and its CO2 content and
that of the true plasma determined. The data of the experiments are
given in Table IX and certain calculations based on these data appear
in Table X. Referring to the latter, several definite conclusions may
TABLE X
Certain Indices of the Distribution of Carbon Dioxide between Corpuscles and Plasma
Specimen No.
19
22
24
25
26
Ratio of CO2 concentration between corpuscles
and plasma
Low pressure experiments
(2.05)
1.01
1.04
1.03
0.82
High pressure experiments
1.43
1.43
1.38
1.29
Change in COs content per unit change in CO2
pressure
Separated serum — vol. per cent per mm.
pressure
0.102
0.093
0.125
0.105
True serum — vol. per cent per mm. pressure.
Whole blood — vol. per cent per mm. pressure
Total buffer value of blood' /3
(0.217)
0.218
9.6
0.127
0.189
7.3
—
0.156
0.208
7.6
0.197
0.234
9.2
Volume of corpuscles — per cent
23.0
34.5
23.00
34.7
31.7
be drawn. In other regards the results of the experiments are at
variance and the interesting question emerges as to whether the apparent
variability in behavior may be due to the properties of the relatively
unspecialized cell which serves as erythrocyte in Urcchis.
The ratio of the CCX concentration of whole blood to that in the true
plasma is approximately 1.0 in the low pressure experiments. The
discordant value in Experiment No. 19 is probably to be attributed to
experimental error. This means that CCX is about equally distributed
between corpuscles and plasma under pressure conditions such as occur
in the blood of the worm. The effects of the Donnan equilibrium and
the corrections for the volume occupied by solutes in the corpuscles and
plasma are neglected and would probably be too small to be significant in
measurements as inaccurate as those employed.
The ratio of carbon dioxide concentration between cells and plasma
in the high pressure experiments is uniformly greater than one. This
result indicates that the principal buffer substances occur within the
corpuscle and that the exchange of materials between corpuscles and
plasma (the chloride shift) which enables the corpuscles to contribute
to the buffer action of the plasma in the higher vertebrates is restricted
in the case of the Urechis blood corpuscle. It has been shown above
that the hemoglobin concentration is sufficient to account for the total
buffer action of Urcchis blood. The apparent restriction in the ex-
200 A. C. REDFIELD AND M. FLORKIN
change of electrolytes between corpuscles and plasma is perhaps to be
related to the tough membrane which may be observed to surround the
erythrocyte of Urcchis.
The relative part played by corpuscles and plasma in Urcchis blood
is further expressed by the estimation of the change in CO., content of
the components per unit change in CO2 pressure in passing from the
low to the high pressure stages of the experiment. This method of
expressing the results is somewhat arbitrary, as the relation is not
strictly comparable for varying ranges of pressure. However, for the
present purpose of comparing data made at two similar pressures con-
siderably separated, it is convenient.
The increase in CO., content with increase of pressure in the case of
the plasma separated at lo-:s tensions is fairly uniform and has an average
value of 0.106 volumes per cent per millimeter pressure. This is almost
exactly the rate of increase which would be due to the solution of carbon
dioxide in the plasma if the absorption coefficient is 0.80 as assumed
above. Parsons and Parsons (1923) publish some measurements of
the carbon dioxide content of sea water at various CO., pressures from
which it appears that the rate of increase is 0.109 volumes per cent per
millimeter pressure. It is concluded that the plasma of Urcchis contains
at most a negligible quantity of buffer material, the increase in CO.,
content being adequately accounted for by the solubility of carbon
dioxide.
Turning to the whole blood, the increase in CO., content with change
in pressure is reasonably concordant in the various experiments and
yields values about twice as great as in tin- case of the separated plasma.
Between the pressures examined the -ain in bound CO.. is about equal
to the gain in CCX dissolved.3 This is a further expression of the fact
that the corpuscular content is responsible for the buffer action of the
blood. The buffer values for the \vhole blood recorded in Table X are
slightly less than that estimated from the experiment recorded in Fig. 5.
The variation in buffer value in different samples of blood does not
appear to be closely correlated with the volume of cells in the samples.
Presumably the variation in hemoglobin concentration in the cells from
different specimens indicated in fable III is sutficient to destroy the
expected correlation.
The true plasma shows a gain in CO.. content with increasing COo
pressure which is variable but always less than the corresponding gain
for whole blood and always greater than the gain shown by plasma
separated at low pressures. The latter fact indicates that with increas-
'•'• < )vrr a shorter range the gain in bound CO- would be relatively greater than
tlii> because <>f tin- " shape " of the carbon dioxide-combining curve.
RESPIRATORY FUNCTION ( )l fK'K( HIS I'.I.OO!)
201
ing CO., pressure some exchange of material between corpuscles and
plasma takes place which increases the ability of the plasma to take up
carbon dioxide. This is presumably a " chloride shift " such as occurs
in mammalian blood. With the exception of Experiment No. 19 the
true plasma always gains less carbon dioxide than does whole blood.
This is a further expression of the fact, brought out in the consideration
of the ratio of carbon dioxide concentration in cells and plasma, that the
exchange of materials affecting buffer action is limited. In Experiment
No. 19 the high value of the carbon dioxide uptake of the true plasma is
probably due to the experimental error which caused the ratio of carbon
dioxide in cells and plasma to appear abnormal.
II. PHYSIOLOGICAL OBSERVATIONS
The O.rygcn Content of the Blood in vivo
Samples of blood were drawn from worms lying in a pan of fresh
sea water by inserting into the ccelomic cavity a hypodermic needle
attached to a graduated one cc. pipette. The blood flowed into the
pipette from its own pressure and was transferred directly into the Van
Slyke apparatus for analysis.4 Immediately following a larger sample
of blood was drawn, equilibrated with air for 20 minutes and analyzed,
thus yielding a measure of the oxygen capacity of the blood. Table XI
TABLE XI
The (\\~ygen Content of Urechis Blood in vivo
Experiment No.
In vivo
Saturated with Air
Temperature
Oxygen Content
Temperature
Oxygen Content
9
°C.
15
volumes per cent
2.87
°C.
18
18
volumes per cent
2.85
2.87
20
18.5
3.45
19
3.70
21
18.5
4.11
19.5
19.5
4.43
4.50
contains the results of three such experiments. The figures are not
corrected for dissolved oxygen. In Experiment No. 9 the oxygen con-
tent of the blood /;/ vivo is equal to that of blood saturated with air.
4 The pressure existing in the coelomic fluid may be about sixteen grams per
cm.2, for on one occasion in drawing blood from the coelomic cavity the blood rose
in the pipette to a vertical distance of 16 cm. and oscillated about this level as the
result of the muscular contractions of the body wall.
202
A. C. REDFIELD AND M. FLORKIX
In the other two experiments the oxygen content is slightly less than
the oxygen capacity. The result indicates that the pressure of oxygen
in the H»od may be considerably less than that of air. The diminished
oxygen content in the blood in vh'o is largely due to the smaller amount
dissolved rather than to incomplete oxygenation of the hemoglobin.
Thus, in the case of Experiment No. 21. if \ve assume the oxygen pres-
sure to be 75 mm., the hemoglobin would be 97 per cent saturated. The
o unbilled oxygen would amount to 3.85 volumes per cent if the total
combining capacity be taken as 3.96 volumes per cent. The dissolved
oxygen would be 0.25 volumes per cent, making the total content of the
TABLE XII
The Carbon Dioxide Content of Urechis Blood in vivo
Specimen No.
In vivo
In vitro
Temperature
CO» Content
O; Contentt
CO: Pressure
CO2 Content
10
°C.
15
;•!./. ff ftnt
7.12
vol. per cent
5.62
mm. Hg
vol. per cent
11
15.5
8.12
4.65
10.4*
36.0*
10.67
18.95
12
16.5
8.79
4.1
See Table VII
i
* Equilibrated at 18° C.
f Equilibrated with air. Xot corrected for dissolved oxygen.
blood 4.10 volumes per cent as observed. It is concluded that the hemo-
globin of Urechis is almost completely saturated when an abundant
supply of oxygenated water is available for respiratory purposes, but
that the pressure of oxygen in the blood may be considerably lower than
that existing in the water.
The Carbon Dioxide Content of the Blood in r'rco
Table XII contains data on the carbon dioxide content of the blood
in z'k'o obtained in a manner similar to that of the oxygen capacity.
The measurements indicate a normal carbon dioxide content of between
seven and nine volumes per cent. Data for the equilibrium of carbon
dioxide with the blood used in Experiment No. 12 are recorded in Table
VI I and Fig. 5. Similar data for two points on the CO.. dissociation
curve of the blood used in Experiment No. 1 1 arc included in Table XII.
These data agree closely with those in Table Y1I. From Table YII it
appears that at the CCX content observed in riro, 8.79 volumes per cent,
the pressure of carbon dioxide would be approximately 7.2 millimeters.
RESPIRATORY FUNCTION OF URECHIS BLOOD 203
77/r /'f/ I'alnc of Urcchis Blood in t'k'o
The pH value of blood plasma is given by the equation
(BHO ' )
in which pK' is a constant dependant upon the properties of the cor-
puscles and plasma. The value of pK' for Urcchis blood is unknown,
but it cannot differ greatly from 6.1 when the corrections for tempera-
ture (Warburg, 1922), ionic strength (Hastings and Sendroy, 1925)
and corpuscular content (Van Slyke, Hastings, Murray and Sendroy,
TRHCO ^
1925) are taken into account. Since the value of log
(H2L(J3)
for Urcchis blood at 7.22 mm. is about one, the pH value of the blood
plasma in vivo must be close to 7.1. The blood is therefore somewhat
more acid than human blood and much more acid than sea water. The
„ 3
of the blood is in solution, the remainder being bound as bicarbonate.
value of log „ 3 indicates that one eleventh of the carbon dioxide
The Exchange of Gas between Blood and Sea Water
Urcchis lives in a permanent burrow in flats which are occasionally
exposed at low tide. The burrows are U-shaped, having two openings.
Water is circulated through the burrow by means of peristaltic con-
tractions of the body wall, which force the fluid backward between the
worm and the wall of the tube. The flow thus established serves both
for respiration and to bring the animal its food supply. The worms
may be kept for long periods in the laboratory confined in artificial
burrows constructed of glass tubing, and under these conditions the
volume of water circulated and the changes in its gaseous content may
be measured. The respiratory and feeding reactions of animals so
confined are fully described by Fisher and MacGinitie (1928). They
consider that respiration is principally effected by means of water
pumped into the hind-gut, through the activity of the muscular cloaca.
The structure of the hind-gut as well as the active rhythm through
which it is ventilated certainly support this view.
The hind-gut is a large sack extending the length of the body and
occupying the greater part of the ccelomic cavity. Its wall is smooth
and so thin as to be quite translucent, resembling in this regard a
mesenteric membrane. The wall of the hind-gut is bathed directly by
the blood, there being no blood vessels (Fisher and MacGinitie, 1928).
The peristaltic movements of the body wall must produce some circula-
tion in the blood. Rhythmic contractions of the hind-gut wall appear
204 A. C. REDFIELD AXD M. FLORKIX
to be important in bringing the blood into contact with the hind-gut wall
as well ns in mixing the water in the hind-gut. If the worm is examined
against the light one may see the outlines of the hind-gut, which appears
as a relatively transparent region. It may be observed that the hind-gut
is the scat of antiperistaltic contractions which sweep over it in the form
i if deep annular constrictions into which the blood is drawn and carried
al'ing. Compared to this effective mechanism the thick, muscular, cutic-
ulated body wall must absorb a relatively small amount of oxygen.
The water within the hind-gut is renewed by a somewhat irregular
rhythm. Fresh water is drawn in by a scries of from one to upward
of thirty small inhalations usually uninterrupted by exhalation. It is
then discharged by means of a single exhalation, frequently followed
by a period of rest. Fisher and Mac( linitie record periods of inhalation
lasting from twenty-five to ninety seconds and expirations consuming
from ten to fifty seconds.
Samples of hind-gut water have been collected at the moment of ex-
piration. The worm is watched through the glass wall of the aquarium
until it begins to discharge the water in a vigorous stream which may be
easily observed. At that moment the worm is taken out of the aquarium
and the anal end thrust tightly into a funnel terminating under oil in a
suitable glass-stoppered bottle. The worm continues to discharge the
hind-gut water, which is collected under the oil. From 25 to 35 cc. of
water may be obtained at a single discharge. Several discharges were
combined to yield material for oxygen analysis by the \Yinkler method.
Carbon dioxide content has been determined with the Van Slyke ap-
paratus. Table XIII contains the results of a number of .such deter-
minations, together with control measurements made upon the sea water
of the aquarium.
T.vuLii XIII
Oxygen and Carbon Dioxide Content of llind-^nt Water
Oxygen
Hind-Rut water 0.37, 0.37, 0.29, 0.36, 0.36, 0.37 vol. per cent
Aquarium water 0.56, 0.56 vol. per cent
Carbon Dioxide
Hind-gut water 5.26, 5.11 vol. per cent
Aquarium water 4.77, 4.79 vol. per cent
The oxygen content of the expired water is about two-thirds that
of the sea water. The partial pressure of oxygen in the hind-gut water
is thus about one hundred millimeters and is quite sufficient to account
for the high degree of saturation found in the blood in z'ivo.
The carbon dioxide measurements mav be evaluated bv means of
RKSriRATORY FUNCTION OF LJRKCHIS BLOOD 205
determinations made by Parsons and Parsons (1923) of the carbon
dioxide content of sea water from the Naples aquarium at various
pressures. They found at 0.8 mm. C'( ), pressure a content of 4.7 vol-
umes per cent which agrees closely with the values found in our aquaria.
Interpolating from their data the carbon dioxide pressure of the expired
water corresponds to 4.6 and 6.0 mm. in the two samples examined.
Since the partial pressure of CO., in the blood is about seven millimeters,
a gradient of pressure of about two millimeters occurs across the hind-
gut wall.
From the foregoing experiments certain deductions may be drawn
relative to the volume of water necessary to "ventilate" the hind-gut.
Dr. V. E. Hall, who has been engaged in a study of the respiratory and
feeding reactions of Urechis, has kindly supplied data concerning the
volume of water pumped by Urcchis through artificial burrows made
from glass tubing, and the rate of oxygen consumption of the worms.
The average rate of oxygen consumption of two medium-sized Urechis
was about 0.013 cc. per minute. The amount of water pumped when
the worms were not engaged in feeding was about 1 1 cc. per minute ;
when feeding, it was about 29 cc. per minute. There is required 2.3 cc.
of sea water containing 0.56 volumes per cent oxygen to yield the 0.013
cc. consumed in one minute. When the water is expired from the
hind-gut only one-third of the oxygen dissolved in it has been consumed.
Consequently 6.9 cc. of water must ventilate the gut each minute. This
is about half the amount pumped through the burrows when feeding
is not going on. Feeding worms pump about four times the required
volume of water, but under these circumstances the water is serving to
bring food to the animal as well as for respiration. The size of the
animals and their activity are variable and consequently these estimations
cannot be very exact. They show, however, that the respiratory activity
of the animal is rather nicely adjusted to the metabolic requirements.
The Function of the Hemoglobin of Urcchis
The data in Table XI show that the hemoglobin of Urechis is almost
completely saturated when an abundant supply of aerated water is avail-
able to the animals. The preceding considerations indicate that the
mechanisms for bringing fresh water into contact with the respiratory
surface of the hind-gut operate with a fair margin of safety at each step.
Under ordinary conditions it appears that the oxygen bound to the
hemoglobin is not utilized and that the oxygen dissolved in the plasma
is sufficient for the metabolic requirements. Urechis must be added to
the list of animals, including Planorbis (Leitch, 1916) and Lianbricns
(Jordan and Schwarz, 1920), in which the hemoglobin does not appear
to function if the oxygen supply is adequate.
206 A. C. REDFIELD AXD M. FLORKIX
Light is thrown on the possible value of the hemoglobin to the worms
by considering the rate at which oxygen " circulates " through the blood.
The problem is somewhat less definite in Urcchis than in the vertebrates
because there are no blood vessels and the ordinary conceptions of
arterial, capillary and venous blood do not apply. If we consider 20 cc.
to be the blood volume of Urechis and four volumes per cent to repre-
-ent the oxygen capacity, then the oxygen content of the total blood is
0.8 cubic centimeters. Taking the rate of oxygen consumption to be
0.013 cc. per minute, it follows that only one-sixtieth of the oxygen con-
tent of the blood is used (and need be replaced) per minute. It is clear
from this why the Urcchis blood is almost completely saturated in vivo.
It also follows that those properties which assist mammalian blood to
give off or take up oxygen and carbon dioxide rapidly during its passage
through the capillaries (the reciprocal action of oxygen and carbon
dioxide on the equilibrium of these gases with hemoglobin and the trans-
fer of buffer action from cells to plasma) may be dispensed with in
Urcchis blood.
The blood of Urcchis appears from the foregoing observations to
contain a store of oxygen sufficient to last the animal one hour. In
addition the hind-gut water itself, having a volume of about thirty cc.,
contains some 0.11 cc. oxygen. This would serve to supply the meta-
bolic requirement for not more than 8.5 minutes. The total oxygen
within the animal consequently is enough to last about seventy minutes.
Consider what would happen if the blood contained no hemoglobin.
In it the oxygen concentration would be no greater than in the hind-gut
water, say 0.37 volumes per cent. The total volume of oxygen in 20 cc.
of blood would be .074 cubic centimeters. At a metabolic rate of 0.013
cc. per minute this would last the animal 5.7 minutes. Adding to this
the time which the oxygen in the hind-gut would serve, the total oxygen
within an animal without hemoglobin would last about fourteen minutes.
The hemoglobin of Urcchis consequently extends the period during
which the respiratory exchange might he interrupted without depriving
the animal of oxygen about five-fold, or for about fifty-live minutes.
This is not long enough to carry the animal over the period of a low tide,
when the burrows are exposed. It is sufficient to be useful during the
" rest periods " which occur after a more or less prolonged period of
feeding. According to Fisher and MacGinitie < T'JS ), these rest periods
are of two sorts: (1) intermittent periods of from 4.5 to 8.5 minutes
separated by about 1.5-minute intervals, during which water is expelled
from the respiratory chamber and a new supply taken. (2) a continuous
-t of an hour or more during which respiration ceases (or at least is
so reduced as to be imperceptible) and no movement of any kind takes
pla
RESPIRATORY FUNCTION OF URKCHIS BLOOD 207
The O.\'\(jen Supply U'lien the Tide is Out
On the California coast the tides follow a rhythm in which alternate
tides are of unequal height. The low course tides are more nearly equal
and in the estuary where Urechis was found the flats are not uncovered.
During the high course tides the flats are uncovered once a day for a
period of six or more hours. During the greatest spring tides the
estuary sometimes empties so completely during the lower ebb tides that
it does not fill during the succeeding flood tide and in consequence the
flats may be bare for 18 hours.
j
During the period when the tide is out there is available for the
worms not only the oxygen in the blood and hind-gut water, which we
have seen is adequate for the metabolic requirements for about seventy
minutes, but also the oxygen dissolved in the water enclosed in the
burrow. An average burrow is about one hundred centimeters long
and two centimeters in diameter. It would contain some 314 cc. of
water, and if this were saturated with air about 1.76 cc. of oxygen.
This would last 135 minutes if used at a rate of 0.013 cc. per minute.
The total oxygen supply of Urechis during low tide is sufficient for only
about three hours according to these calculations.
In order to throw more certain light on the state of affairs during
low tide, a series of analyses on the oxygen content of the water in the
burrows was made. A rubber tube was thrust down into the burrows
and sufficient water for analysis by the Winkler method (70 cc.) drawn
out and transferred to a glass-stoppered bottle without exposure to air.
\ new burrow was selected for each observation. The flats had al-
ready become bare when we arrived but had not been so for more than
one-half hour to judge from the state of the tide when the first ob-
servation was made. The last observations were made from the last
burrows to be covered after the greater part of the flat was submerged
by the rising tide. The oxygen content of the water left in a puddle by
the receding tide, which serves to give an idea of the content of the bur-
row water before the flat was bared, was 0.34 volumes per cent. This
relatively low value may be accounted for by the fact that the observa-
tions were made at daybreak. The water had been overnight in an
estuary teeming with animal and vegetable life and oxygen losses had
not been compensated by photosynthesis. The temperature of the water
in the burrows was 15° C. at 6 :08 A.M. and had risen to 17° C. at 11 :30.
At this time the water in the channel was 19° C. The results are re-
corded in Table XIV.
During the first hour after the flat is bare the oxygen content of the
burrow water appears to decrease rapidly and at about the rate expected
from the foregoing calculations. There is some irregularity in the
208
A. C. REDFIELD AND M. FLORKIX
TABLE XIV
Oxygen Content of Water from Urechis Burrows During Low Tide
Approximate Time
Oxyeen Content of Water
Became Bare
From Individual Burrows
Mean
r, OS \.M
hours
fl ^
cc. per 100 cc.
fl 71
cc. per
100 cc.
fl 71
n 7
0 Ifi
fl 1 6
6-30
fl 0
0 1 }
fl 1 3
6:35
1 0
015-011-011
0 P
7 :4C)
7 n
(I U
fl ?4.
7 :55
> >
0 1 6
fl 1 6
8:00
) ^
0 1 ~>
fl 1 7
9:30
A n
0 06- fl OfV fl 06
0 06
10:30
S 0
0 16- o 14- () 14
0 14
11:15
s 7
0 06- 0 16- 0 "> V 0 '7- 0 ~>1
0 ?0
individual measurements made during the third hour, hut hy the fourth
hour the oxygen content has definitely sunk to a minimal value of 0.06
volumes per cent. During the fifth hour there is a perfectly definite
increase in the oxygen content of the water in almost all of the burrows
examined. These measurements support the view that the oxygen in
the water inclosed within the burrow and in the blood is insufficient to
maintain the normal metabolic rate for the duration of the low tide.
After the first hour the oxygen in the burrow water diminishes rather
slowly and one must conclude, either that the rate of oxygen consump-
tion by the worm is diminished or that the oxygen in the burrows is
replenished hv some means. There is reason to believe that both these
processes occur. It is well established that the metabolic rate of many
marine organisms varies with the oxygen pressure in the environment
(Amberson, Meycrson and Scott, 1924; Hall, 1929, and others). The
measurements made on the water of the- burrows very definitely indicate
an increase in the oxygen content during the last hour before the flats
were covered. This suggests that the water in the burrows is slowly
replaced by the water with which the sand is permeated. The effect
is probably related to changes of hydrostatic pressure within the flat
occasioned by changes in the tide level, for it is reported that wells in
sandy soil near the sea sometimes display definite changes in level related
to the tides. The effect becomes noticeable onlv during the last hour
when the tide is rising rapidly. It is presumably occurring throughout
the low tide period and selves to check the exhaustion of the oxygen
content of the water by the metabolism of the worms. If ibis view is
correct it serves to explain how i'rccliis can withstand the 18 hours of
RKSP1RATORY FUNCTION OF URKCH1S I'.LOOD
200
low water which occur during the- spring tides. At these times the
small intermediate tide, although unable to cover the flats, will serve to
move about the water within the flats and thus replenish to a certain
degree the oxygen within the burrows of Urcchi.t.
The oxygen content within the burrow never appeared less than
0.06 volumes per cent, which corresponds to an oxygen pressure of
about fourteen millimeters. At this pressure the hemoglobin of Urccliis
is nearly 60 per cent saturated. During the greater part of the low tide
the pressure of oxygen in the burrow is such that the hemoglobin of the
blood will function effectively as an oxygen carrier while very little
oxygen will be present in solution in the blood. Provided the oxygen
in the burrows does not sink below the observed levels, the hemoglobin
of the blood may be expected to transport an adequate supply of this gas
to the organs of the body.
SUMMARY
1. The blood of Urccliis caitpo contains hemoglobin enclosed in cor-
puscles. The oxygen capacity of the blood varies from 2.66 to 7.22
volumes per cent and the percentage of cells in the blood from 18 to 40.
2. The oxygen dissociation curve is measured. Its position does
not appear to be influenced by the carbon dioxide pressure. The effect
of temperature upon the oxygen dissociation curve is of the direction
and order observed in other bloods containing hemoglobins.
3. The carbon dioxide dissociation curve is measured. The ability
of the blood to combine with carbon dioxide does not appear to be in-
fluenced by the degree of oxygenation of the blood.
4. The buffer value of the blood is 11 volumes per cent and is con-
stant over a considerable range of carbon dioxide pressures. The con-
centration of hemoglobin accounts for the entire buffer effect.
5. Carbon dioxide is about equally distributed (in concentration) be-
tween the corpuscles and plasma. The plasma contains at most a
negligible quantity of buffer material. With increased carbon dioxide
tension there is a small, but distinctly limited exchange of material be-
tween the corpuscles and plasma which increase the ability of the latter
to combine with carbonic acid.
6. The hemoglobin in vivo is almost completely saturated, but the
pressure of oxygen in the blood may be considerably less than that in the
surrounding water. The carbon dioxide content /// vivo is 7 and 9 vol-
umes per cent, corresponding to a carbon dioxide pressure of about
seven millimeters Hg. The reaction of the blood is estimated to be
about pH 7.1.
7. The " ventilation '' of the respiratory organ, the hind-gut, is
210 A. C. REDFIELD AND M. FLORKIN
considered quantitatively, the result indicating that the respiratory ac-
tivity is nicely adjusted to the metaholic requirements.
8. The function of hemoglohin and its relation to the oxygen supply
during low tide are discussed. It is suggested that the movement of
water within the flats due to changing tidal level is important in sup-
plying oxygen when the tide is out.
REFERENCES
AMBERSON, \V. R., H. S. MF.YERSOX, AND \V. T. SCOTT, 1924. Jonr. Gen. Ph\s'wL,
7: 171.
BARCROFT, J., AND H. BARCROFT, 1924. Proc. Roy. Soc.. Ser. B., 96: 28.
BOHR, C, 1897. Ann. de Physik. it. Chemie, 62: 644.
BROWN, W. E. L.. AND A. Y. II in.. W23. Proc. Roy. Soc.. Scr. B. 94: 297.
DILL, D. B., AND H. T. EDWARDS, 1931. Jour, Biol. Chan., 90: 515.
DKASTICH, L.. 1928. Compt. rend. Soc. de Bwl, 99: 991.
FISHER, W. K., AND G. E. MACGINITIE, 1928. Ann. and Mag. Nat. Hist., Ser. 10,
1 : 199 and 204.
Fox, H. M., 1926. Proc. Roy. Soc., Ser. B, 99: 199.
HALL, F. G., 1929. Am. Jour. Physiol., 88: 212.
HALL, F. G., AND I. E. GRAY, 1929. Jour. Biol. Client.. 81: 589.
HARI. P., 1917. Biochem. Zeitschr., 82: 229.
HASTINGS, A. B., AND J. SENDROY, JR., 1925. Jonr. Biol. Client.. 65: 445.
HEXDERSOX, L. T., 1928. Blood, A Study in Gem-pal 1'hvMolt^y. New Haven.
JORDAN, H.. AND B. SCHWARTZ, 1920. Pfliiger's Arch., 185: 311.
KENNEDY, R. P., 1926-27. Am. Jour. Physiol., 79: 346.
LF.ITCH, I., 1916. Jour. Physiol.. 50: 370.
MACELA, I.. AXD A. SELISKAR, 1925. Jour. Physiol., 60: 428.
PARSONS. T. R., AND YV. PARSONS, 1923. Jour. Gen. Physiol., 6: 153.
SKIT/.. 1007. Zoo!. Jahrhuch., Abt. Anat.. 24: 323.
VAN DER Li- GI \. J.. AND L. HOGBEN, 1928. Trans. Roy. Soc. S. Africa. 16: 2"5.
VAN SI.VKK. h. 1).. A. I'.. HASTINGS, C. 1). MTRRAY. AXD J. SEXDROY. JR., 1925.
Jour. Bio!. Client.. 65: 701.
VLKS. F., 1923. Arch. Phys. Biol., II, No. 6.
\VIXTERSTEIN, H., 1909. Biochem. Zeitschr., 19: 384.
OXYGEN AND CARBON DIOXIDE TRANSPORT BY THE
BLOOD OF THE URODELE, AMPHIUMA TRIDACTYLA
WALTER J. SCOTT
(From the Department of Physiology, and the l)ef>(irliuent of Research Medicine,
School of Medicine, University of Pennsylvania)
This paper is a presentation of the oxygen dissociation curves and
of the carbon dioxide absorption curves of the blood of Amphiuma
tridactyla together with comparisons with similar curves from the litera-
ture of the carp, the turtle, the frog, and man. In particular, the ap-
plicability of Hill's equation to the bloods of these species and the shape
of the oxygen dissociation curve as an adaptive mechanism are briefly
discussed. In addition, the properties of the carbon dioxide absorption
curves, especially of those which are a result of the low hemoglobin con-
tent, are brought out.
TECHNIC
The blood was drawn from the heart 3-5 minutes after the injection
of one cc. of 1 :1000 heparin in 0.7 per cent sodium chloride to prevent
clotting of the blood in the syringe during withdrawal. It was then
transferred to a glass container coated with sufficient sodium fluoride and
sodium oxalate to make a final concentration of about 0.1 to 0.2 per cent
and kept at 2-8° C. until used.
The equilibration of the blood with the desired tensions of oxygen
and of carbon dioxide, for 30 minutes at 22-26° C., was performed by a
technic similar to that of Austin, Cullen, Hastings, McLean, Peters, and
Van Slyke (1922). Blood gases were determined in duplicate by the
method of Van Slyke and Neill (1924). Analyses of the gas phase
were made by the Haldane-Henderson apparatus. Hematocrite deter-
minations of corpuscular volume were made before and sometimes after
samples were drawn for equilibration. No difficulty was encountered
in the use of the ferricyanide method for Amphiuma blood in contrast
with the experience of Krogh and Leitch (1919) with fish blood.
Caprylic alcohol as an antifoamer was omitted because of clot formation.
OXYGEN CONSUMPTION OF AMPHIUMA BLOOD
The spontaneous diminution of the oxygen of blood so common
particularly with nucleated cells is quite marked in Amphiuma blood.
It was found (Table I) that keeping the blood at 2-5° C. except during
211
212
WALTER J. SCOTT
the period of actual equilibration practically eliminated this spontaneous
oxygen consumption. The presence of physiological amounts of carbon
dioxide apparently diminishes appreciably the oxygen loss. The slight
loss which may occur during 20-30 minutes of equilibration and hand-
ling is negligible. \Yastl (1928) has successfully used KCX to prevent
oxygen consumption in the blood of the car]), but its effect in Ainpliinina
blood appeared variable and we abandoned its use in favor of cold to
eliminate this source of error.
T \HI.E I
Effect of Temperature on the Oxygen Consumption of Ampkiuma Blood
Temperature = 20° C.
Time
minutes
0
Oxygen Content
i<ol. per cent
2 64
59
1.41
118
0.20
197.
. 0.0
Temperature = 3° C.
t >\\ U' •!) ( '' nit i-ii t :it Y.irviir..'. ( '( )• Tension-
Time
ninnttes
1-3 mm.
25 nun.
42 mm.
44 mm.
5? mm.
60 mm.
0
3.49
4.64
3.00
4.57
4.52
6.23
20
2.98
4.52
6.18
60
3.16
4.70
2.64
4.41
4.46
5.94
120
2.89
4.61
2.55
4.23
4.19
180
2.69
4.50
2.69
3.99
240
2.69
1.5')
Till'. Sol.ri'.ll.li Y C '()K K I- H IK NT OK <>.\Y<;K.\ IN . \ M I' I I I r M A 1>LOOI>
It is usually assumed as a sufficient approximation that tin- solubility
of a gas in blood or serum, relative to its solubility in water is propor-
tional to the water content of the blond. Fur Amptihinia blood the
relative solubilities in serum and in whole blood accordingly would be
about (>5 per cent and <S/ per cent respectively. \Ye have measured
CtCX directly in both scrum and whole blood of Ainpliiniiia in the fol-
lowing way: Samples of serum and of whole blood were equilibrated
with air and with ()(>.6 per cent oxygen. To diminish spontaneous
oxygen consumption the equilibration was performed at 4 C. and the
solubility o efficient relative to that of water at the same temperature
calculated. After equilibration blood was drawn into a mercury re-
ceiver. Sampling and transfer to the Van Slyke apparatus was ac-
complished by using a Barcroft pipette. Assuming that the oxygen of
KK.sriK.vi K >\ ( >!• .\\irinr.\i \
213
the air is sufficient to fully saturate the hemoglobin, aO.2 may he cal-
culated from the difference in the oxygen content of the samples by the
equation :
where V and V are respective oxygen contents in volume percentage
and P and P' are respective oxygen partial pressures in mm. 1 Ig. Table
II shows that the relative solubilities of oxygen in Ainphiitina serum and
TABLE II
Solubility Coefficient of Oxygen in Amphiuma Whole Blood and Serum
Temperature
Oxygen Tension
Oxygen
aOs
0 C. ± ..5°
3
mm. Hg
749
157
I'd. per cent
3.90
0.84
0.0393
6
754
158
3.98
0.79
0.0408
3
Serum
3
754
748
157
3.98
4.04
0.83
0.0401
0.0403
4
157
0.84
0.0410
Av. 4
4
751
157
Average aOv
765
161
3.97
0.82
= 92% of «Oo water
12.23
9.05
0.0403
0.0400
4
764
161
12.28
9.05
0.0406
Whole blood 4
740
156
12.90
9.77
0.0398
Av. 4
756
159
Average aO2
12.45
9.29
= 92% of aO2 water
0.0402
I
whole blood are about 92 per cent of that in water ; and this figure is
used in the subsequent calculations. These figures fail to demonstrate
a difference in the aO., between serum and whole blood of Aniphiuma,
which is not surprising in view of the low corpuscular volume of the
blood.
15
2\4 \VALTKR J. SCOTT
HEMOGLOBIN CONTEXT OF THE CELLS
The corpuscular volume varies considerably and is slightly higher
than that found hy Southworth and Redfield (1926) for the turtle.
The oxygen capacity also varies greatly in Aphinma hlood. In Table
II I it is seen that the ratio of oxygen capacity to cell volume is roughly
TABLE III
Variation in Oxygen Capacity and in Corpuscular Volume in the Blood of Amphiuma
and the Ratio of Oxygen Capacity to Cell Volume
Rr-rl Colls Oxypen Capacity Oxygen Capacity
vcl. per cent vol. per cent cell volume
14 ..................... 2.52 0.18
15 ..................... 4.52 0.30
17 ..................... 4.7 0.28
20 ..................... 5.01 0.25
20 ..................... 5.8 0.29
25 ..................... 6.1 0.24
28 ..................... 8.38 0.30
30 .................... 9.64* 0.27
52 ..................... 9.42* 0.18
* Concentrated hlood.
constant, and averages 0.25. The data of Southworth and Redfield
(1926) on the turtle show an approximate ratio of oxygen capacity to
cell volume of 0.50 and the a \erage figure for human hlood is said to
he 0.45. The red eells of both Ainphhtnia and the turtle are nucleated
and as a rough approximation of the fraction of the volume occupied
by the cytoplasm \ve take 0.80. and calculate the concentration of hemo-
globin per unit volume of cytoplasm. It appears from these data that
the concentrations of hemoglobin in the cvtoplasm ot the red cells of
Amphiuma and the turtle are 0.32 and 0.63 volumes of oxygen per
volume of cytoplasm respectively, or 71 per cent and 140 per cent of
the concentration of hemoglobin in the cytoplasm of the human red cell,
an interesting divergence in this factor f.n- the three different species.
Tin-: < ).\Y<;K.\ DISSOCIATION CURVE
The oxygen-binding properties of both hemoglobin solutions and
bloods, except at low and high degrees of saturation, may be repre-
sented with sufficient accuracy by the equation of Hill (1910) :
lib
\Ve have found it convenient to analyze our data on Amphiuma
blood by means of this equation and it is evident that it holds quite well
for this blood as shown by Fig. 1. Here are plotted the points of an
RESPIRATION OF AMPHIUMA
215
oxygen dissociation curve at 43 ± 3 mm. of carbon dioxide (Table IV)
and for comparison, points from similar data for the carp (Wastl, 1928),
turtle (Southworth and Redfield, 1926), and man (P>ock, Field and
. \dair, 1924). The respective curves arc calculated from the values of
70
c
«
o
60
150
0 40
1O
10
I-Corp blood at 1&°C. and -SOmm.CO, fenaion. O
H- Amphiuma • • Q.6"C. - 43*3 •• A
HL-Tur-tle, - - 25"C. • 4O Q
EZ'-Muman " •• 38X1 4O - • •
10
1O 25
O x. y tf fj n
50
FIG. 1. Oxygen dissociation curves. The lines are calculated by Hill's
equation and plotted using the ;; and A"0 values (Table V) from the original data
(points).
/; and K,, for the blood of each species, given in Table V. These con-
stants in turn were obtained in the usual way by plotting log HbO2/Hb
against log PCX, as in all cases, the points between approximately 10 to
90 per cent saturation fall quite closely on a straight line represented by
the equation,
HbO,
log -~ - log A,, + n log P,
from which the constants AT,, and n have been calculated. The con-
formity of the experimental observations (points) with the equation of
Hill (lines) shows that Hill's equation holds reasonably well within the
limits specified for these divergent species. Ani^hiuma resembles
human blood in showing an " S ' shaped oxygen dissociation curve
which is absent in the carp and turtle. It is well known that the
' S " shaped curve is associated with high values of N (greater than
1.8). This is apparent from the comparison of the curves of these four
216
WALTER J. SCOTT
TABLE IV
Oxygen Dissociation Curve of Amphiunin Blood. Temperature = 26° C.
Carbon dioxide tension = 43 ± 3 mm. Oxygen capacity == 8.4 vol. per cent.
«O2 = 0.92 aO2 (in water).
O: Tension
HbOj
Saturation
mm. Hg
vol. per cent
per cent
1.
2.2
0.01
0.0
2.
8.4
0.70
8.3
;
15.8
2.2
26.3
4.
26.2
3.9
47.0
5.
56.4
6.7
80.0
6.
79.6
7.6
91.0
7.
143.5
8.4
100.0
forms whose ;/ values vary from 1.4 to 2.0. The absence of the " S "
shape has been interpreted by Krogh to signify an adaptation of bottom
forms such as the carp to low oxygen tension, whereas the free-swim-
ming forms, e.g., the trout, whose curves are " S " shaped. d<> not possess
this adaptation. The air-breathing Amphiunia, despite its habit of re-
maining submerged for considerable periods, appears not to possess
this adaptation. That the " S " shaped curve is not a necessary property
• it" the blood of air-breathing animals is evident from its absence in the
turtle, whose n value is 1.5.
TABLE V
Hill's n and K0for Carpi, Turtle, Man, and Amphiutnu l\'ln>lc Blood
ies n A'o X 10-»
Carp 1.3
Amphiuma 1.8
Turtle 1.5
Man. 2.0
36.0
2.5
8.3
1.9
Wast I (1928)
Southworth and Redfield (1926)
Bock, Field, and Adair (1924)
EFFECT OF CARBON DIOXIDK ox TIIK OXYC.F.N DISSOCIATION CURYF.
The increased acidity of the blood with increased carbon dioxide
tension has been abundantly shown to decrease the oxygen saturation
at a given tension. In other words. A.',, is decreased by increasing carbon
dioxide tension. That this effect is manifest in . •Iniphiitniti is shown
by the data of Table VI. The values of A',, have been calculated by
Hill's equation, using n 1 .X. This decrease of the affinity of hemo-
globin for oxygen with increased acidity is in accord with experience in
many species such as man, clog, fishes, and turtle.
RESPIRATION OF AMPHIUMA
217
TABLE VI
Effect of Carbon Dioxide on Ka of A mphiuma Blood
1.
Oxygen Capacity
CO2 Tension
Oxygen Tension
Saturation
A'010-3
vol. per cent
5.8
mm. II g
a. 2.3
mm. Hg
53.0
per cent
88.0
5.7
b. 38.0
83.0
84.0
1.8
2.
4.5
a. 1.2
26.0
58.0
3.9
b. 43.5
85.0
87.0
2.2
3.
2.5
a. 1.9
44.0
78.0
4.2
b. 42.7
48.0
62.0
1.5
THE TRANSPORT OF CARBON DIOXIDE
Figure 2, made from the data of Table VII, shows three carbon
dioxide absorption curves of oxygenated Aphnmia blood. Curve I is
characterized by a rather low CCX capacity, about 39 volumes per cent
at 70 mm. CCX tension, and also by the fact that when this blood is
equilibrated with gas mixtures lacking carbon dioxide, very little of this
gas remains in the blood. Curves II and III, on the other hand, are
80
70
f-
2 60
U
U
Of 50
U
a
u 40
D
30
U
a
X
o
a
z
o
<Q
Ck
u
20
10
10
16 % CELLS
13% CELLS
34% CELLS
100
20 30 40 50 60 70 80
CARBON D i OXIDE: TENSION mm.Hq
FIG. 2. Carbon dioxide absorption curves of oxygenated Amphiuma blood.
Temperature 24 + 2° C.
218
\YAI.TKR T. SCOTT
characterized by a much higher carbon dioxide capacity, approximately
63 volumes PIT cent at 78 mm. carbon dioxide tension, and further by
the fact that when these bloods are exposed to low carbon dioxide tension
during equilibration a relatively large amount of carbon dioxide, 30-40
volume's per cent, still remains in the blood. The considerable difference
between curve I and curves II and 1 1 1 is probably due to the large
variation in hemoglobin content. It is well known that hemoglobin
functioning a> an acid can combine with only a limited amount of base
so that onlv limitc-d amounts of XallO).. will be decomposed into
sodium hemoglobinate and carbon dioxide when blood which contains
but a limited amount of hemoglobin is exposed to xero or low tension
of carbon dioxide.
TAUI.I-: \ II
Carbon Dioxide Absorption Curves. (See Fig. 4.) Temperature =: 24 ± 2° C.
Oxygenated blood.
Curve Xo.
Point N"<>.
COj
i . i
I (Cells = 34 per cent)
1.
2.
mm. //i;
1.1
6.7
;'(>/. per tent
4.48
14. OS
3.
16.9
22.36
4.
34.4
30.25
5.
70.3
38.85
II (Cells =• 19 per cent)
1.
3.4
39.45
2.
12.9
46.11
3.
19.5
48.79
4.
45.4
56.78
5.
76.1
62.94
1 1 1 (Cells - 18 per cent)
1.
1.1
28.62
2.
16.3
43.0
3.
45.5
53.32
4.
76.4
M 05
The amount of sodium bicarbonate which will be decomposed by the
hemoglobin in pa-sin- from a definite tension, e.g. 40 mm. of carbon
dioxide, to a tension of zero may be calculated as follows: let the
maximum base bound per unit of hemoglobin be btl and the base bound
at 40 mm. carbon dioxide per unit of hemoglobin be /'. Then if | lib]
be the concentration of hemoglobin in the blood, the XaMO), decom-
posed in the reaction—
Xaiuxx + nub — \aiii. i- n,ma
is obviously A XalICO, (/'„--/')• I llb.| A rough extrapolation "t
the curves of Hastings. Sendroy and I leidelberger (1924) for horse
RESPIRATION OF AMPHIUMA
2\()
hemoglobin gives .7>,,---S and b==2. That is, blood in passing from
40 mm. to zero millimeters of carbon dioxide tension will decompose no
more than l\,--b volumes of Nal K'< ).. for each volume per cent of
oxygen capacity. For the Amphluma bloods whose carbon dioxide
absorption curves are shown in Fig. 2 the oxygen capacities are re-
spectively 8.5, 4.5 and 4.5 volumes per cent. The corresponding
amounts of NaHCO;. decomposable are therefore 51, 27 and 27 volumes
per cent. In the case of curve I, however, the amount: of NaHCO3
at 40 mm. was 32 volumes per cent which, being less than the maximum
of 51. will be entirely decomposed at zero carbon dioxide tension. This
was found to be the case as shown in Fig. 2. In the case of curves II
and III, where at 40 mm. carbon dioxide tension the NaUCCX is ap-
proximately 55 volumes per cent, only 27 volumes per cent will be de-
composed. This was found to be approximately the case as seen in
Fig. 2. This calculation, of course, is admittedly the roughest sort of
approximation from insufficient data and is offered only as a semi-
quantitative explanation of the phenomenon of incomplete decomposition
80
70
60
h
2
U
u
tk
u
a 5°
40
x
O
Q 20
O
g
10
u
TURTLE
FRO&
AMPHIUMA
HUMAN
CARP
10
20
30
40 50 60
DIOXIDE TENSION
70
80
100
FIG. 3. Carbon dioxide absorption curves of carp, frog, Amphiuma, turtle
and man. Temperature, human curve 38° C., all others, room temperature.
220
WALTER J. SCOTT
of XaHCO;; by hemoglobin at zero millimeters of carbon dioxide ten-
sion. A quite similar qualitative explanation of the same phenomenon
in turtle blend was first offered by Southworth and Redfield ( 1926).
60
U
Ck
u
a «o
u
D
O 30
U
Q
Q
2
O
CO
U
ZO
10
10 20
CAR OON
30 40
Diox > o E
80
50 60 70
TENSION mm.Ha
Fir;. 4. The effect of oxy^enation and reduction on the carhon dioxide ab-
sorption curve of Amphiuma blood. Kiuhteen per cent cells. Temperature
24° C.
Figure 3 shows a number of carbon dioxide absorption curves for
the carp, frog. Ainphhtnia. turtle, and man, representative of four of
the classes of vertebrates. All of the curves for the lower vertebrates
resemble each other more than they do the mammalian curve. In these
few examples the amphibia occupy an intermediate position with respect
to carbon dioxide content between the teleostian carp and the reptilian
turtle. The marked flatness of the turtle curve is attributed by South-
worth and Redfield (1926) to the low corpuscular volume, i.e.. the low
lib of turtle blood. \Vastl (1928) gives the same sort of explanation
for tli' 'tially parallel curve for carp blood. A similar explanation
for the flatness of the Amphiuma curve is indicated just as it is lor the
other lower vertebrates, since the blood of all of these animals shows
a corpuscular volume only about one-half to one-third that of human
blood.
RESPIRATION OF AMPHIUMA
221
THE EFFECT OF OXYGENATION AND REDUCTION ()!•• THE BLOOD ON THE
CARHON DIOXIDE TRANSPORT
Figure 4 shows the difference in the amounts of carbon dioxide
carried by oxygenated and reduced .Iti/p/iiitnm blood. The difference
in carbon dioxide content in the two cases is not so great for Amphiuma
blood as that found by Christiansen, Douglas, and llaldane (1914) for
human blood. In fact, this difference amounts to about two volumes
TABLE VIII
The Effect on Carbon Dioxide Content of Oxygenation and Reduction of Amphiuma
Blood. (See Fig. 4.) Temperature = 24° C.
Curve No.
Point No.
CO» Tension
CO"
I (Reduced blood, O2 capacity 4.4 vol. per
cent)
1.
2.
mm. Hg
0.7
22.7
vol. per cent
30.87
47.28
3.
39.8
56.94
4.
43.6
54.34
1 1 (Oxygenated blood, Oa capacity 4.4 vol.
i;
1.1
28.62
per cent)
2.
16.3
43.0
3.
45.5
53.32
4.
76.4
61.05
per cent at physiological levels for Ainphinina, while for man the cor-
responding value is about 5.5 volumes per cent. If the difference in the
amounts of carbon dioxide carried by oxygenated and by reduced blood
be divided by the oxygen capacity of the sample of blood, a ratio — ^=-,
i.e., the increase of carbon dioxide content per unit of oxygen capacity,
is obtained. The value of this ratio in the case of human blood is
about 0.28 volume of carbon dioxide per unit volume of oxygen capacity.
In five experiments we attempted to determine this ratio closely, but
our results were quite divergent. \Ye found for Amphiuma blood
AGO,
values of — — p- as follows: 0.23, 0.31, 0.46, 0.71, and 0.93, averaging
0.54. It is obvious that the calculation of this ratio is subject to con-
siderable error since it is the ratio of small differences of large volume ;
nevertheless, the results are all in the same direction as in the blood of
man. The mean value of the ratio for Amphiuma is, of course, very
approximate. Physiologically, however, oxygenation and reduction
have little effect on the transport of carbon dioxide by Amphiuma blood.
In this respect, also, the Anipliiunta is like the turtle. The data for the
curves of Fig. 4 are included in Table YIII.
\YALTKR J. SCOTT
SUMMARY
1. The corpuscular volume of Amphiuma blood varies considerably,
from 12 to 35 per cent.
J. The oxygen capacity varies from 3 to 10 volumes per cent.
3. It is >ho\vn that the oxygen dissociation curve is like the typical
mammalian curve with certain features in common with those of the
turtle and carp.
4. The presence of physiological amounts "f carbon dioxide affects
the oxygen dissociation curve in the usual way.
5. The comparative values of ;/ and A.',, of the 1 1 ill equation are given
for the blood of + linf>hiunui. carp, turtle and man, and the equation of
Hill for these bloods is shown to hold within the limits specified.
6. The mechanism for the transport of carbon dioxide in the
Amphiutna blood is much like that in the turtle and the flatness of the
carbon dioxide absorption curve is explained as a function of the limited
amount of hemoglobin.
7. The difference in carbon dioxide carried by oxygenated and re-
duced blood is quite small and probably has little physiological signifi-
cance, though the increase in carbon dioxide content per unit of oxygen
capacity is in the same direction as that for man.
The major portion of this work lias been carried out under the
direction of Dr. William C. Stadie. i wish to thank him for his interest
in this problem and for his continual aid and encouragement.
BIBLIOGRAPHY
AUSTIN, I. II.. (.. E. < ULLEN, A. I'.. HASTINGS, F. C, McLEAN, J. P. PETERS,
ANI. I). I). VAN SLYKE, l'»22. Jmir. liiol. Chem., 54: 121.
BOCK, A. V., M. FIELD, JK.. AND (',. S. . \I..\IR. 1924. Jour. Jiiol. Chan.. 59: 353.
BOHR, CHRISTIAN, l''i>5. SL;m,l. Arch. f. I'liyxiol.. 17: 104.
BOIIK, ('.. K. H ASSKI.HAI.CM. AMI A. KKIICH, 1904. Sk<iii<1. Arch. j. J'liysioi..
16: 402.
( IIKISTIANSF.X. J., C. '.. Doll, I. As, \\|l I. S. II AI.DA \K. 1(M4. Joitf. PIlVSUll.,
48: 244.'
HASTINGS, V I!.. FULII s Si-.\in«>v. (.'. I). MTKRAY, AND MKIIAKI. HEIDKLBF.RGER,
1^24. Jour, liiol. Chem.. 61: 317.
HILL, A. V., 1910. Jour. Phys'wl., 40: iv-vii, Proceedings of 1'MO.
KI«H;H, A., AND I. LKITCH, 1919. Jmtr. I'liysin!.. 52: 288.
Soi "i H WORTH, F. C., JR., AMI A. (\ RF.DKJKI n. l''_'(i. Jour. Gen. Ph\sioL. 9: 387.
VAN SLYKE, I). I).. AND I. M. XKII.L, 1^24. Jour. liiol. Client.. 61: 523.
WASTI, II., l'»2H. Him-hcm. Zcitschr.. 197: 363.
WASTL, H., AND A. SKI.ISKAR, 1925. Jour. Ph\swl.. 60: 264.
MOVEMENT AND RKSl'nXSK IX I)IFFLIT<;IA WITH
S1MCUAL REFERENCE TO THE NATURE OF
CYTOPLASMIC CONTRACTION' '
S. O. MAST
THE JOHNS HOPKINS UNIVERSITY
INTRODUCTION
It has been repeatedly observed by various investigators that when
Difflugia and other shelled rhizopods travel, pseudopods, one after an-
other, extend in a given direction, become attached at the tip to the
substratum, then shorten and pull the shell forward ; but the only ref-
erence concerning the mechanism involved in these processes is found
in a former paper in which I came to the following conclusions (1926,
p. 413) : " In this process of locomotion the tip of the attached pseudo-
pod functionally becomes the posterior end. The plasmagel probably
changes into plasmasol here and then flows directly into the new pseudo-
pod. . . . The extension of the pseudopods is ... dependent upon con-
traction in the plasmagel, resulting in local pressure on the plasmasol."
However, in this work but little evidence in support of these conclusions
was obtained from observation on Difflugia. They were largely based
upon the results of detailed observations on the process of locomotion
in Amoeba. Fortunately, I have recently had the opportunity, under
very favorable conditions, to make equally detailed observations on the
process of locomotion in Difflugia. These observations are considered
in the following pages.
MATERIAL AND METHODS
Two species of Difflugia were used in this investigation: D. f>yri-
fonnis (Leidy) and D. acuiniiiata (Leidy). Both were found in the
ooze on the bottom of a large permanent pond, the edges of which were
frequented by cattle and horses. Pyrlfonnis was abundant, acuiuinata
rather scarce. The pond is located near Town Hill, Mt. Desert Island,
Maine.
All of the specimens studied were well filled with Chlorclla.
1 Contribution from the Mt. Desert Island Biological Laboratory. I am
greatly indebted to the Director of this laboratory for excellent laboratory facil-
ities and to the Research Corporation for financial aid in procuring assistance in
the investigation.
223
224 s. O. MAST
They were very active and they lived well in the laboratory, both in jars
and on slides under cover-glasses sealed with vaseline. They were
consequently very favorable for making extensive observations on
locomotion.
The process of locomotion was studied under Zeiss apochromatic
objectives and compensating oculars with magnifications ranging from
_'i >() to 1200 diameters. With the lower magnifications the specimens
were observed in watch-glasses, with the higher under cover-glasses sup-
ported and sealed with vaseline. In some observations the cover-glass
was far enough from the slide so that the specimens could move about
freely; in others it pressed on the shells of the difflugiae just enough to
prevent locomotion, and in still others it pressed on the shells so much
that they broke, some slightly, others considerably. After the shells
were broken some of the specimens left and moved about naked. The
process of locomotion was thus studied in specimens with shells and in
specimens without shells.
The response to tactile and photic stimulation was also briefly studied
as indicated below.
LOCOMOTION
*
Difflugia pyriformis m'ith Shells Free
The shell of Difflugia pyriformis is flask-shaped but usually con-
siderably flattened. It consists of a layer of sand grains which vary
greatly in si/.e. The interstices between the grains of sand are filled
with a yellowish substance which holds them together. According to
Leicly, the shells vary greatly in size, ranging from .06 to .58 mm. in
length and from .04 to .24 mm. in width. In the specimens studied the
shell was about .4 mm. long and .2 mm. wide.
When at rest the living portion of Diffluyia is usually entirely within
the shell and it fills only about three-fourths of the space. When it
begins to move a pseudopod forms on the surface of the body below the
opening in the shell, then extends out through the opening and advances
free into the surrounding medium, until it is about as long as the shell
(Fig. 1). As the pseudopod advances it usually swings from side to
side freely and extensively but not rapidly. The tip thus frequently
moves through an arc of nearly 90° in a little more than one second.
In this swinging the tip sooner or later comes in contact with the sub-
stratum to which it adheres. Then it contracts slowly and pulls the
shell along. Before this pseudopod has disappeared another one usually
develops at the opening of the shell and extends at a considerable angle
with the old one, but as the old one disappears, the new one becomes
directed forward, attaches and then contracts and pulls the shell for-
MOVEMENT AND RESPONSE IN DIFFLUGIA
ward. This process is repeated, one pseudopod developing after an-
other and each pulling the shell forward .2 to .4 mm. Movement of
the shell is consequently intermittent.
Sometimes the new pseudopod attaches before the old one detaches
sb
FIG. 1. Camera drawing of Diffluyia pyriformls in locomotion, sh, shell
constructed of sand grains cemented together; />, pseudopods ; /, plasmalemma ;
g, plasmagel ; s, plasmasol ; arrows, direction of flow in the piasmasol ; c, Chlorclla;
a, disc of cytoplasm projecting to the edge of the .-hell ; nun., scale showing mag-
nification. The stippling of the plasmagel in this and the following figures is not
intended to represent structure. The plasmagel contains numerous small granules,
but the plasmasol contains an equal number of the same description.
The large region almost entirely within the shell, enclosed by the solid line,
contained so many chlorellae that it was dense green. This region was covered
with a thin layer of hyaline substance. The pseudopods usually contain but few
chlorellae. At the mouth of the shell the hyaline layer ( /; ) usually extends out
over the edge of the shell forming a sort of cushion.
Note that the plasmasol extends to the tip of the pseudopod and that there is
no hyaline cap.
and occasionally two pseudopods extend simultaneously, become attached
and contract simultaneously (Fig. 2C). Pseudopods also sometimes
branch at the tip or near it and elsewhere. When a pseudopod is ex-
tending or advancing there is rapid streaming forward in the middle.
This becomes slower and slower toward the surface and disappears
entirely before the surface is reached, i.e., there is a layer immediately
below the surface which does not move forward. This layer is usually
very thin and it is evidently relatively solid, i.e., gel. It forms a tube
through which the more fluid part, the plasmasol, flows. At the tip of
the pseudopod, the plasmasol stream spreads out and as it comes in
contact with the end of the plasmagel tube it gelates, i.e., it is here trans-
formed into plasmagel. The plasmagel tube is consequently built for-
ward by gelation of plasmasol at the tip of the pseudopod just as it is
in Aniccba. But the plasmagel tube is usually open at the tip of the
pseudopod and the plasmasol flows forward to the end with nothing
corresponding to the hyaline cap in Amoeba (Fig. 1).
226 S. O. MAST
The plasmagel and the plasmasol in the pseudopod contain innumer-
able small granules and after the pseudopod is fully extended, a group of
chlorellae usually appear in the plasmasol at the base. These are,
however. never carried to the tip where the plasmasol is transformed
into plasmagel, and they consequently never get into and become a part
of tin- plasmagel (Fig. 1).
The surface of the pseudopod is covered with a very thin membrane
which is in fairly close contact with the plasmagel. This membrane
cannot be seen directly, but the fact that the plasmasol flows to the
very tip of the pseudopod and stops there, and the fact that hyaline
blisters form in various regions on the surface, as will be demonstrated
presently, show that there is a surface membrane, a plasmalemma. This
doubtless slides over the plasmagel and stretches as the pseudopod ex-
tends. The pseudopod in Difflugia is therefore point for point per-
ceptually the same in structure and in the process of extension as it is
in Amoeba.
Nothing could be directly ascertained concerning the mechanism in-
volved in the extension of pseudopods in Di/jhitjid. but the fact that they
are not in contact with the substratum when they advance shows that
they are pushed out by contraction of the portion of the body in the
-hell. This contraction is doubtless in the plasmagel. just as it is in
The bending of the pseudopod from side to side is doubtless due to
local contraction of the plasmagel on the side of the pseudopod toward
which it Ix-nds. This contention is strongly supported by the results
obtained in observations on response to contact, presented in a succeed-
ing section of this paper.
As soon as the tip of the pseudopod comes in contact with the sub-
stratum it adheres to it. Then it flattens and spreads over the sub-
stratum and as it spreads it attaches. This continues until the attached
surface has increased three or four times in width. Thus the' tip of the
pseudopod becomes very firmly fastened to the substratum (Fig. 2).
Immediately after the pseudopod has become attached, one or more
blisters form at the point of adhesion to the substratum. These consist
of droplets of fluid which apparently have been squeezed out of the
plasmagel. The fluid aggregates between the outer surface of the
plasmagel and the substratum and then spreads laterally (Fig 2/>).
This fluid is definitely differentiated from the surrounding medium.
showing that there must be on the pseudopod a surface membrane, a
plasmalemma, which has been separated from the plasmagel by the fluid
squeezed out. Similar blisters are formed under other conditions as
indicated below.
MOVEMENT AND RESPONSE IN DIFFLUGIA
22;
When such blisters form, fluid can actually be seen to flow from the
plasmagel, which now becomes clearly visible as a thin granular layer.
Under certain conditions this granular layer can be seen to break after
the blister containing hyaline fluid has been formed (Fig. 2A, Z>) ; then
the granular plasmasol flows through and disperses throughout the
hyaline fluid. In this fluid there are a few scattered granules which
h—
1
I
sh
0.1
mm
r
FIG. 2. Camera outlines of pseudopods of Diffluyla fyrlfonms, illustrating
the process of attachment. A and B, first stages in attachment; C, attachment
complete; h, hyaline blisters; s, plasmasol; f/, plasmagel; I, plasmalemma.
Note that during the process of attachment the hyaline layer in the region
that becomes attached increases greatly in thickness and spreads out over the
substratum and then gelates.
In the specimen represented in C two pseudopods extended and attached simul-
taneously. This does not often occur. In C the pseudopod attached some distance
back of the tip.
are in violent Brownian movement. These facts show that the hyaline
substance in the blister has liquid properties and that the plasmagel
layer is relatively solid. After the blisters have fully formed and
have spread over the substratum the Brownian movement in them
ceases, indicating that the fluid in them has gelated. This conclusion
is supported by other evidence presented below.
S. O. MAST
Attachment of the pseudopod must he due to adhesive substance
nn the MIT face of the plasmalemma. or to adhesive character of the
plasmalemma itself. The fluid in the blister formed immediately after
the pseudopod becomes attached must be squeezed either out of the
plasmagel in the region of adhesion, owing to local contraction, a process
resembling syneresis ; or out of the plasmasol. owing to local increase
in the- water permeability of the plasmagel and contraction in this layer
elsewhere. The spreading out on the substratum and the flattening of
tlu- tip must be due to surface tension which pulls the edges of the
pseudopod in all directions over the substratum in the same way that
the edges of a drop of oil on water are pulled by surface tension over
the surface of the water. If this obtains, the surface tension of the
pseudopod-water interface plus that of the pseudopod-substrat interface
must be less than the surface tension of the substrat water interface.
After the tip of the pseudopod is attached, the plasmagel becomes
thicker and every portion of the pseudopod shortens, resulting in marked
and fairly rapid decrease in the length of the pseudopod and in con-
siderable increase in thickness. During the process of shortening of
the pseudopod the plasmasol usually does not flow toward the shell as
rapidly as the plasmagel retracts, i.e., the plasmasol actually flows for-
ward in relation to the plasmagel. although it is moving toward the shell.
Sometimes, however, a branch forms near the tip of a pseudopod while
it is contracting and when this occurs the plasmasol usually actually
flows forward, i.e., it flows forward in reference to the shell and in ref-
erence to poinN outside the organism. This shows that retraction of a
pseudopod in DijjlHfiia differs considerably from retraction of a pseudo-
pod in Amoeba. In the latter, as I have demonstrated elsewhere ( 1926),
the pseudopods shorten, owing to transformation of plasmagcl into
plasmasol at the tip. In the former they shorten owing to contraction
of the plasmagel throughout the entire length. This is, however,
eventually followed bv transformation of the plasmagel into plasmasol
at the tip. Retraction of the contracted pseudopods in Difflugia is there-
fore like the retraction of the pseudopod in Amccba.
Sometimes pseudopods are fully extended and then retracted without
having become attached. When this occurs retraction is usually much
more rapid than it is when they are attached and after they have
shortened considerably numerous small blisters appear scattered over the
surface, except near the tip; but they become more and more abundant
as one approaches the base of the pseudopods. Yerworn ( 1889) ob-
served the formation of similar blisters during retraction of pseudopods
in Difflugia urceolata. These blisters increase in number as the pseudo-
pods decrease in length (Fig. 3). They are, as will be demonstrated
presently, associated with thickening of the plasmagel.
MOVKMKNT AM) K I- M'< >\SK L\ U1FFLUGIA
229
After a pseudopod has become attached, the fluid in the blisters in
contact with the substratum gelates and the plasmagel throughout the
entire length of the pseudopod thickens. This is, however, much more
FIG. 3. Camera outlines of pseudopods of Difflugia pyrifonnis, illustrating
contraction without attachment to the substratum. A, pseudopod fully extended;
B and C, stages in contraction.
Note that during contraction numerous hyaline blisters form on the surface.
This is due to localized thickening of the hyaline layer. The thickness of the
plasmagel increases markedly when the hyaline blisters form.
evident in specimens in which the shell is firmly held by the pressure of
the cover-glass than it is in those in which the shell is free. I shall
consider this matter more fully in the following section.
Difflugia fiyriforinis with Shells Fastened to the Substratum
The shell of Difflugia was fastened and held as follows: Several
specimens which were nearly the same in size were mounted in water
under a cover-glass supported with a ridge of vaseline. Then water
was slowly removed until the cover-glass pressed upon the shells just
enough to prevent locomotion but not enough to break the shells. In
specimens in this condition all the processes of locomotion observed in
specimens with the shells free occur except movement of the shell, but
there are some illuminating modifications in some of the processes.
Extension, bending and attachment of pseudopods are in all respects
precisely the same under the two conditions, but the result of contraction
16
230 S. O. MAST
is very different, the shell moving forward under the one condition and
the attachment of the pseudopod hreaking under the other.
After the pseudopod has hecome attached in specimens with the
cover-glass resting on the shell, the plasmasol gradually hecomes
narrower and streaming in it, slower; the plasmagel becomes thicker,
and the entire pseudopod some little distance hack of the point of at-
tachment becomes thinner. The portion of the pseudopod between the
point of attachment and the shell becomes perfectly straight and nu-
merous lines running parallel with the longitudinal axis of the pseudopod
appear in the plasmagel, especially near and in the region of attachment
(Fig. 4). All this indicates marked strain, owing to violent contraction
of the plasmagel especially in the proximal region of the pseudopod.
This continues with incrca>ing force until the attachment gives way.
This sometimes takes place suddenly. If it does the pseudopod shortens
so rapidly and so extensively, after the attachment has been broken,
that the distal end actually snaps back to a point not more than half as
far from the shell as it was. This demonstrates conclusively that the
pseudopod was under rather violent strain before the attachment broke.
Usually the attachment of the pseudopod breaks gradually. When this
obtains, one point after another, here and there throughout the entire
attached portion of the pseudopod, gives way and owing to this the
regions of the plasmagel at the tip, which do not give way, are drawn
out in strands of considerable length. Finally the attachment of these
also breaks, but they retain their form for some time, giving the tip of
the pseudopod a distinctly fibrous, brush-like appearance. These
strands gradually retract, but the distal end of the p-eudopod remains
very irregular in outline and much flattened until it disappears (Fig.
4 C. D, E, F).
The facts that much of the substance in the attached portion of the
pseudopod is drawn out in strands, that these1 strands contract rapidly
and extensively after attachment to the substratum breaks, that the
•/
anterior surface of the pseudopod retains an irregular contour until it is
withdrawn, and that the anterior end of the pseudopod remains much
flattened after it is free, demonstrate conclusively that this substance is
a fairly linn, highly elastic gel. These facts and others presented in-
dicate that contact induces gelation, resulting in increase in thickness of
the plasmagel throughout the entire length of the pseudopod, and that
this causes increase in the elastic strength of the plasmagel, resulting in
contraction in the pseudopod and expansion elsewhere. JHit why does
the substance contract after it gelates? This is obviously the central
problem concerning the processes involved. It is easy enough to under-
-land whv and how an elastic substance which has been stretched con-
MOVEMENT AND RESPONSE IN DIFFLUGIA
231
III
FIG. 4. A series of camera sketches of a pseudopod of Diffluyia pyriformis
representing different stages in the process of contraction in a specimen with the
shell fastened to the substratum. A, early stage in the attachment of the pseudopod
to the substratum ; B, pseudopod firmly attached at the tip and beginning to contract
elsewhere. Note that the plasmagel has increased in thickness and that it appears
to be fibrous at the tip. C, D, and E, stages in the contraction of the pseudopod.
Note that as the pseudopod contracts, and the attachment breaks, the cytoplasm
at the tip of the pseudopod is drawn out into irregular strands which usually
shorten considerably after the}' become detached, indicating that the cytoplasm
here is viscous and highly elastic. This was particularly evident in the projection
labeled .r. The tip of this projection remained attached to the slide for some
time after the rest of the tip of the pseudopod became free. This resulted in
great stretching of the projection. Finally the attachment broke, whereupon the
projection contracted very rapidly and extensively as indicated. F, outline repre-
senting the cross section of the pseudopod near the distal end. Note that the
pseudopod was much flattened.
tracts but why and how a sol which has gelated and has not been
stretched contracts is difficult to understand. I shall consider this prob-
lem presently.
Difflugia pyriformis without Shells
Difflugia without a shell is, as far as my experience goes, never found
in nature. Sometimes it will, however, leave its shell if the shell is
slightly broken, as the following observations indicate.
232 S. O. MAST
Six specimens varying considerably in size were mounted under a
cover-idass supported and sealed with a small ridge of vaseline. Then
the cover-glass was gently pressed down until the shells were broken.
si line verv slightly, others considerably, after which observations were
made from time to time for six days. During the first day there was
practically no movement in any of the specimens in which the shell was
considerably broken. They were much rounded and there was no
indication of formation of pseudopods. but they did not disintegrate.
Those in which the shell was only slightly broken behaved normally,
i.e., pseudopods extended, became attached and contracted precisely as
they do in specimens with intact shells. ( >n the second day pseudopods
appeared I nun time to time on various sur laces in those with badly
broken shells. These pseudopods were- at first short, but later they
extruded practically as far as they do in normal specimens. The for-
mation of these- pseudopod.s was not related to the mouth of the shell.
They apparently developed equally readily on all surfaces, extending
here and there through crevices in the broken shell. They often ap-
peared alternately on opposite sides, one extending while the other
contracted.
Later in the dav three of the specimens left the shells. One. how-
ever, carried with it the edge of the month of the shell in the torm of a
ring. Through this psendopods extended, one after another, became
attached at the tip and contracted, pulling the body along just as in
normal specimens. Xo pseudopods tunned elsewhere on the body. In
this specimen it could be clearlv seen that the plasmasol in the contract-
ing psendopod (lowed directly into an extending pseudopod which de-
veloped from tin- base of the contracting pscudopod as a branch. There
wa.s. however, no .such violent and rapid shortening oi pseudopods as
was sometimes seen in normal .specimens, especially when stimulated, as
will be shown later. The shell therefore seems to function in this.
Moreover, the shell serves to coordinate movement in that it contmes the
formation of pseudopods to one region on the surface of the body;
namely, that opposite the opening in the shell. This becomes evident
if the movement of normal specimens or that of the specimen with the
ring, just described, is compared with that of the specimens which were
entirely naked.
In these specimens after they had left the shells just as before they
had left it, the formation of psendopods was not restricted to one sur-
face. As a matter of fact, succcssj\r pseudopods rarely lormed in the
same region of the bodv ; indeed they often formed on diametrically
opposite surfaces. This resulted in movement, now in one direction and
MOVKMKNT AND RFSPOXSK IX DIFFLUGIA
233
then in another; movement which in comparison with that of normal
specimens was very irregular in direction and quite uncoordinated.
For some time after these specimens left the shells the pseudopods
extended, attached and contracted, pulling the body along, much as they
do in normal specimens. Sometimes two pseudopods appeared on op-
FIG. 5. Camera sketch of a specimen of Difflugia pyrifonnis two days after
it had left its shell which had been broken by the pressure of the cover-glass. In
this specimen two pseudopods which extended in opposite directions were several
times seen to attach simultaneously and then to contract. This resulted in great
elongation of the specimen. The portion enclosed by the broken line was well
filled with chlorellae; the rest contained numerous small grayish granules.
posite sides of the body simultaneously; then extended, attached and
contracted, pulling the body out in opposite directions and greatly
elongating it (Fig. 5). The following day all of the specimens were
out of the shells and there were several small ones. These were prob-
ably fragments which had separated from the large ones. A large
234
S. O. MAST
granular nucleus and several contractile vacuoles could now be clearly
seen in each of the large specimens (Fig. 6), but none was found in the
small 01
Tlu- small specimens moved about in a fairly coordinated fashion
!^r. 7) and some of the large ones now moved much more consistently
in a given direction than they did on the preceding day, and in these the
Fi<;. 6. Camera sketch of a specimen of niplnt/ia f>yriforntis several days
after it had left its broken shell, n. nucleus; i\ contractile vacuole; arrows,
direction of flow in the plasmasol.
This specimen moved about fairly consistently and the process of locomotion
was essentially the same as that in .-lunrba frotcus. It contained many chlorellae,
but they were scattered, making it possible to see the nucleus and the contractile
vacuoles.
process of locomotion was in all essentials like that in Amccba protcus
(Mast, 1926). Attachment of the pseudopods at the tip followed by
contraction had practically disappeared. A pseudopod advanced in a
given direction for a time, then stopped but did not contract. In t be-
meantime another appeared near its base, advanced in the same general
direction for a time and stopped, etc.. just as in Ainceba protcits.
The movement of the plasmasol in the pseudopods could be very dis-
tinctly seen in these specimens, owing to the fact that the chlorellae
were carried out in the pseudopods very much farther than they were in
normal specimens. Sometimes they were carried to the tip of the
pseudopods, but they never were caught in the gelation of plasmasol at
MOVEMENT AND RESPONSE IN DIFFLUGIA
235
the tip and consequently never became a part of the plasmagel. This
resulted in a very definite differentiation between the plasmasol and the
plasmagel, the former being green and the latter greyish.
The following day movement continued in the same way but the
specimens were less active, and 24 hours later all of the specimens were
I*— 0.1 mm — >\
Vic,. 7. Camera outlines of a fragment of Dlffluyia fyriftinnis produced by
cutting- off a pseudopod from a normal specimen. Interval between successive out-
lines, 2 minutes; arrows, direction of streaming.
Note that the fragment changed rapidly in form and that relatively large
pseudopods developed. This fragment moved about freely and the process of
locomotion was in full accord with that in Amoeba protcus. The fragment
contained a surface membrane (plasmalemma) , a thin gel layer under this
(plasmagel), and a central fluid mass (plasmasol).
much rounded and there was but little movement. All of the specimens
lived a few days more, i.e., they lived without shells nearly a week.
These observations were repeated twice with essentially the same
results. The fact that the process of locomotion in naked specimens
of Difflugia is like that in Amoeba protcus strongly supports the conten-
tion that in normal specimens the principles involved are fundamentally
the same as in Aunvba.
Difflugia acnminata
Difflugia acnminata is very much like Difflugia pyriformis in struc-
ture, the only marked difference being a protuberance on the base of the
shell; but it is much smaller, the specimens studied being only about .17
mm. in length and .1 mm. in width.
The process of locomotion is essentially the same in the two species.
In both, pseudopods one after another form, extend, attach and contract,
pulling the shell along in steps. In both the plasmagel in the pseudo-
pod is in the form of a tube open at the end, and the plasmasol flows out
236 S. O. MAST
through this ami gelates at the distal end. thus building the gel tube for-
ward and extending the pseudopod. In both, when the pseudopod con-
tracts, the plasmasol flows back through the plasmagel tube, but in D.
itciiniiiuita the backward flow is much more regular than in H. pyriformis
and it almost invariably flows directly into a new pseudopod which
u>ually extends as the old one contracts ( Fig. S). while in /'. pyriforuiis
it usually flows back into the body and from there out into a new
• -endopod.
Locomotion in P. acuminatceis essentially in accord with the descrip-
tion of locomotion in Difflugia sf>. presented in a preceding paper (Mast,
1926. p. 413). It resembles locomotion in .-Inucba protcus with the ex-
ception of the rather violent contraction of the pseudopod after attach-
ment at the tip. resulting in marked periodicity in the rate of movement.
There is. however, also a tendency toward periodicity in the rate of
movement in .Inncha protcus which was clearly demonstrated by
Schwitalla (1924), and it may be that the processes involved in produc-
ing this periodicity are fundamentally the same in both forms, i.e., that
there is a certain amount of contraction in the pseudopod after attach-
ment in slina'l'u as well as in Difflugia.
FIG. 8. Camera outline of l)iffl;i>/i<i aciiiuiunlii, illustrating the flow of
inasol from the old PM udopod directly into the now one.
TO T.ICIIT
ia p\>-ifonnis is definitely positive to light. It does not orient,
at least not preciselv, but in dishes lelt in tront o! a window it ag-
gregates on the more highly illuminated side. This was repeatedly ob-
-erved in the culture dishes, but it was more evident in a series of tests
made by evenly distributing specimens in watch glasses in front of a
north window, leaving them for a time and then ascertaining the
distribution.
The results obtained in several tests are essentially the same. In
one test consisting of five watch-glasses containing a total ot 131 in-
dividuals there were at the end of 12 hours 103 in the window-half of
I lie dishes and only 29 in the opposite half.
MOVEMENT AND RESPONSE IN DIFFLUGIA
237
Concerning- the process of aggregation I have no information.
Rapid increase in illumination probably causes cessation in streaming,
but this response if it actually does occur is far less definite than it is in
.Innrba profcns. In naked specimens it was observed that locomotion
is much more rapid in low illumination than it is in high, but rapid in-
crease in illumination did not result in sudden cessation in streaming.
r\
/
V B N
FIG. 9. Camera outline of Diffluyia pyriformis, illustrating bending of pseudo-
pods. A, bending without stimulation; B, bending after light mechanical stimula-
tion at x; 1, original position of pseudopod; 2, position 1.5 seconds later; a,
position of pseudopod before stimulation; b and c, positions one and two seconds
later respectively; d, blister formed on the side of the pseudopod at the point
stimulated.
RESPONSE TO CONTACT
Response to contact was studied as follows : Several specimens of
Difflugia pyriformis were put into a watch-glass under a dissecting
binocular and left until they became active. Then with a glass needle
pseudopods in different stages of development were touched in various
ways and the effects noted. The results of numerous observations may
be summarized as follows :
If the needle is gently brought into contact with the side of a pseudo-
pod, the pseudopod at the point touched almost immediately bends
sharply and fairly rapidly toward the side touched. The tip, in this
response, frequently swings through more than ninety degrees in less
than two seconds (Fig. 9). This bending is obviously due to contraction
S. O. MAST
of the plasmagel in the region stimulated, Here, immediately after the
needle comes in contact with the surface, there is formed a small blister.
In a preceding section we have noted that the formation of blisters is
associated with thickening of the plasmagel, i.e.. with gelation of
plasmasol adjoining the plasmagel in the region where the blister ap-
pears. If this is true, it is evident that contact causes thickening of the
plasmagel, owing to gelation of adjoining plasmasol and that this causes
local increase in the elastic strength of the plasmagel and contraction in
this region resulting in the bending of the pseudopod.
It. in place of bringing the needle gently in contact with one spot on
the pseudopod, it is in close succession brought rather violently in con-
tact with several spots fairly uniformly distributed over the surface, the
entire pseudopod contracts, and this usually continues until the pseudo-
pod has been drawn entirely into the shell. The contraction of the
pseudopod is sometimes gradual, the pseudopod gradually receding into
the shell, but it is usually very rapid, as a matter of fact, so rapid that
the pseudopod jerks back into the shell much as a tubicolous annelid
jerks into its tube when it is violently stimulated. Yerworn (1889,
1914) obtained similar responses in Difflugia urccolata.
The withdrawal of the pseudopod is often retarded by a mass of
cytoplasm in the mouth of the shell, from which the pseudopod projects.
This mass of cytoplasm forms a sort of stopper; it tills the mouth of
the shell and projects over the edge in the form of a flange (Fig. 1).
When this obtains, the pseudopod, after it is stimulated, shortens, then
holds its position until the flange gives way. after which the whole mass,
pseudopod and all, suddenly darts into the shell.
The fact that this whole mass of cvtoplasm is thus suddenly drawn
into the shell shows that stimulation of the pseudopod causes marked
contraction in that portion of the body which is located in the neck of the
shell and that the body of the organism is fastened to tin- shell in the
basal region; for if this portion of the body were not fastened, it would
be drawn forward in place of the portion in the neck of the shell being
drawn backward.
When a pseudopod, owing to stimulation, contracts as described
above, numerous small hyaline blisters torm on the surface. This, as
previously demonstrated, is associated with thickening of the plasmagel,
owing to gelation of the adjoining plasmasol, and this in turn producer
increase in the elastic strength of the thickened plasmagel, resulting in
contraction. The fact that stimulation of the pseudopod causes not only
gelation and contraction of the region stimulated but also ot certain por-
tions of the body within the shell. I.e., in a region some distance from the
location of the stimulus, shows that localized contact stimulation pro-
MOVEMENT AND RKSl'ONSE IN DIFFLUGIA 239
cluces in Difflugia something which is transmitted through the cytoplasm
and then causes gelation of plasmasol followed l>y contraction. That is,
it produces something which is akin to what in higher forms is known
as an impulse. This conclusion is supported l>y results obtained hy
Verworn in observations on the effect of localized mechanical stimulation
of Difflugia urccolata. He (1889, 1914) maintains that local stimula-
tion of a pseudopod causes, under certain conditions, contraction of
pseudopods which were not stimulated. There is, however, no evi-
dence indicating that this obtains for all rhizopods. Verworn was not
able to find it in other species of Difflugia and it probably does not occur
in Amceba (Mast, 1932).
DISCUSSION
I have demonstrated in the preceding pages that the more important
factors involved in locomotion and response in Difflugia consist of sola-
tion of the plasmagel in one region of the body and gelation of the
plasmasol in another, of attachment of the tip of the pseudopod to the
substratum, of gelation of the plasmasol in the pseudopod correlated
with contact and of contraction of the pseudopod due to increase in the
thickness and in the elastic strength of the plasmagel in it.
All of these factors are probably also involved in the process of
locomotion and response in Aniccba, but gelation associated with contact
followed by contraction of the pseudopod, which plays such an important
role in the process of locomotion in Difflugia, is of little if any signifi-
cance in the process of locomotion in Amoeba. To account for locomo-
tion in Difflugia, it is necessary then to explain not only gelation and
solation but also contraction of the substance after it has gelated. Con-
cerning the processes involved in this, I have no suggestions to offer
except that the fact that fluid is squeezed out during gelation and con-
traction indicates that the processes involved are in some respects similar
to those associated with syneresis as found in the gelation of various in-
animate substances.
SUMMARY
1. The fleshy part of Difflugia pyriforniis and Difflugia acuinlnata
is in structure much like Amoeba proteus. There is a thin elastic sur-
face membrane (plasmalemma), a central fluid mass (plasmasol), con-
taining a large granular nucleus, and a relatively solid layer (plasmagel)
which surrounds the plasmasol. Probably there is also a hyaline fluid
layer between the plasmagel and the plasmalemma but this, if it is
present, is much thinner and less conspicuous than in .-Inurba.
2. Locomotion in Difflugia is normally brought about by the exten-
sion of pseudopods, one after another, and attachment to the sub-
240 S. O. MAST
stratum at the ti]>. followed by contraction which pulls the shell con-
taining the body forward. Movement is consequently intermittent.
3. The elastic strength of the plasmagel is lowest at the tip of the
pseuclopods. This results in contraction of the plasmagel elsewhere, and
this contraction forces the plasmasol nut through the plasmagel tube,
causing expansion at the tip of the pseudopod. The plasmasol which
is in contact with the distal edge of the plasmagel tube continuously
-elates and this results in extension of the tube.
4. After the tip of the pseudopod becomes attached, the plasmasol in
the tip gelates and the plasmagel throughout the entire pseudopod
thickens greatly owing to gelation of adjoining plasmasol. This in-
creases the elastic strength of the plasmagel in the entire pseudopod
until it become- greater here than elsewhere, after which it contracts
and the pseudopod becomes shorter and thicker and the plasmasol in it
is forced back into the 1« >d\ of the organism.
5. The extension of pseudopods in Difflugia is in principle the same
a- in .luia'ba protcns, and contraction is probably also the same in prin-
ciple, but it is much more pronounced and much more highly specialized
in Diffliif/'ui than in .//mr/></. in which it does not function appreciably in
the proofs of locomotion.
<>. The pseudopods in Difflugia alter they are extended wave about
considerably. This is doubtless due to unequal local contraction ot the
plasmagel on opposite sides.
7. If the shell of nif}h(</ia pyrifonuis is broken, it leaves the shell in
the course of a day or so and moves about fairly freely. Specimens out
of the shell sometimes live for a week or more under a cover-glass sup-
ported and scaled with a ridge- of vaseline. After they have been out of
the shell for some time the process of locomotion in such specimens is
in all respects like that in . iiiucini, although it is usually much more
irregular in direction. Contraction ot extended pseudopods, so con-
spicuous in the process of locomotion in specimens containing shells, has
practically disappeared.
(S. Difflni/iu [>\'>-ii(»'niis aggregates on the more illuminated side of
dishes in moderate illumination. It does not orient precisely. It is less
active in high illumination than in low. especially when it is out of the
shell. It may be that this functions in the aggregation observed. Rapid
increase in illumination probably causes decrease in the rate of stream-
ing, but this, if it occurs, is much less marked than it is in .-liua'ba
proteus.
9. Weak local contact stimulation of an extended pseudopod causes
-harp bending in this region toward the side stimulated. Strong general
contact stimulation causes rapid contraction of the entire pseudopod.
MOVEMENT AND RESPONSE IN D1FFLUGIA
The bending is clue to local thickening of the plasmagel in the region
stimulated. Contraction of the entire pseudopod is due to thickening of
the plasmagel in the entire pseudopod.
10. Contact stimulation results, under certain conditions, in gelation
which extends far beyond the region stimulated. There is therefore in
Diffluyia transmission of something akin to an impulse in higher forms.
LITERATURE CITED
MAST, S. O., 1926. Structure, Movement, Locomotion and Stimulation in Amoeba.
Jour. Morpli. mid Physio!.. 41: .547-425.
MAST, S. O., 1932. Localized Stimulation, Transmission of Impulses and the
Nature of Response in Amoeba. Pliysiol. Zool., 5, in press.
VERWORX, MAX, 1889. Psycho-phsiologische Protisten-studien. Jena, 218 S.
YKRWORN, MAX, 1914. Erregung und Lahmung. Jena.
THE EFFECT OF DILUTION OF SEA WATER OX THE
ACTIVITY AXD LONGEVITY OF CERTAIN .MARINE
CERCARLE, WITH DESCRIPTIONS OF
TUX) NEW SPECIES
II. \V. STUNKARD AND C. RUTH SHAW
(I- row the Biological Laboratory, New York University, and the Marine Biological
Laboratory, Woods Hole, Mass.)
INTRODUCTION
The present investigation was undertaken to secure data bearing on
the question of the origin, distribution, and evolution of present groups
of digenetic trematodes. The specific problem under consideration de-
velops from the observation that several families of the digenetic trema-
todes have representatives in both marine and fresh-water hosts. A
brief statement of the problem was given by Stunkard (1930). If the
trematodes found in both marine and fresh-water hosts and assigned to
common families, and even to common genera, have true phylogenetic
relationships; i.e., if they have descended from common ancestors,
rather than consisting of groups that through convergence show mor-
phological and developmental similarities, their distribution raises an
exceedingly difficult biological problem.
There appears to be little doubt but that the parasites in question are
actually closely related. Among the gasterostomes, Bucephalus poly-
tnorphns von Baer, 1826 was described from fresh-water fishes and
Bucephalus haimeanus Lacaze-Duthiers, 1854 was described from
marine fishes. Tennent (1906) traced the life cycle of the latter species,
and the recent studies of Woodhead (1929, 1930) have demonstrated
the development of two species that occur in fresh-water hosts. The
similarity in structure and development between the marine and fresh-
water species is so striking that it strongly indicates close relationship.
In the Prosostomata there are a number of families whose members
occur in both marine and fresh-water hosts. The family Aspido-
gastricke, e.g. (see account by Stunkard, 1917), contains species that
infest mollusks, fishes and turtles of frrsh water, and others that occur
abundantly in marine fishes. In this aberrant family, also, the mor-
phological and developmental agreement is too close to be satisfactorily
explained on the basis of convergence.
Several other families of the Digenea manifest the same type of
242
DILUTION OF SEA WATER AND CERCARI^ 243
distribution. The family Fasciolidae (see Stunkard and Alvey, 1930)
contains one group of genera which infests the livers of terrestrial
herbivores, and another, consisting of Campida, Lecithodesinus, Ortlio-
splanchnus, and ZalopJwtrcuia, which occurs in tlie livers of various
marine mammals. In the family Paramphistomida; there are a large
number of genera, most of which parasitize the hoofed mammals, al-
though one genus, Chiorchis (see Stunkard, 1929) occurs in the Atlantic
manatees. It is of course possible that the sea cows, frequenting the
mouths of rivers, acquire these parasites in fresh water. With few ex-
ceptions, members of the Heterophyidae occur in terrestrial vertebrates
and the larvae develop in fresh-water snails, while Cr\ptocotyle lingua
infests the intestine of fish-eating birds and the larva? develop in the
marine snail Littorina llttorca (see Stunkard, 1930(?). In the family
Pronocephalidae, several genera have been reported from marine turtles,
while one species was described (Stunkard, 1930) from the fresh-water
turtle, Amy da, and a second has been described by Mackin (1930), from
Pseudciuys clcguns. According to Fuhrmann (1928) the family
Steringophoridae (syn. Fellodistomidse, see Stunkard and Nigrelli, 1930)
contains species from the intestine of both marine and fresh-water
fishes. Representatives of all of these groups have been studied by the
senior author and the results afford cumulative evidence that both
marine and fresh-water hosts harbor closely related species of parasites.
Nicoll (1915, 1924) lists other families which have representatives
in both marine and fresh-water fishes. Furthermore, there are several
genera, e.g., Azyyia (see Manter, 1926), which have species in both
marine and fresh-water hosts. No exhaustive review of the literature
is here attempted, but sufficient data have been presented to indicate that
many groups of trematodes which manifest marked similarity in struc-
ture and development have members, some of which infest marine and
others which infest fresh-water hosts. The agreement in morphology
and manner of development, recurring so consistently in different
groups, can hardly be fortuitous, and the majority of investigators are
agreed that these groups are formed by closely related rather than con-
vergent species.
If, as has been postulated, these groups contain closely related
species, the question naturally arises as to whether the common ancestral
form occurred in marine or fresh-water hosts and how the present dis-
tribution was effected. A factor which must be kept in mind through-
out the discussion is the complicated life history of these digenetic forms.
Typically, sexual multiplication occurs in a vertebrate host, and eggs are
produced which pass from the body of the host. From the eggs there
emerge aquatic, ciliated larva? which invade the first intermediate host,
244 H. W. STUNKARD AND C. RUTH SHAW
always an invertebrate and usually a mollusk. where asexual multiplica-
liun takes place. A second type of aquatic, tailed larv;e leaves the first
intermediate host, and these larva-, either by direct penetration or after
encyst meiit mi aquatic plants or in the bodies of other intermediate hosts.
final lv reach the vertebrate host. The free-living, larval stages are ex-
tremely delicate, ephemeral, and incapable of any extended migration.
( 'onscqueiitly the presence of members of a common group in both
marine and fresh-water hosts can only he explained by migration of the
Imsts or by transfer to new hosts.
The migration of free-living species from a marine to a fresh-water
habitat, or the reverse, is largely prevented by the physical, chemical, and
biological factors that eharactcri/e the two types of environment. It is
true that marine species have been cut oil in arms of the ocean, e.g., the
Caspian and l.lack Seas, which have subsequently become bodies of
fresh water, and some of them have persisted although the number of
Mich species is not large. Among the fishes, the anadromous and
catadromous forms make regular migrations from one habitat to the
other, but these examples stand as exceptions to the general rule. Fresh
water imposes an effective barrier against the migration of Foraminifera.
corals, echinoderms, cephalopods, and other groups ot invertebrates.
The relatively few invertebrates that have transferred from the ocean
to fresh water have undergone extensive modifications in form and in
life history. The free-swimming larval stages, characteristic ot marine
types, have almost entirely disappeared.
In an excellent study of this subject Xeedham (1930) has discussed
the factors which prevent the penetration of marine organisms into
fresh water. Allee ( 1(>23 ) has shown the effect of differences in tem-
perature, oxygen content, and hydrogen ion concentration on the dis-
tribution of littoral invertebrates. Adolph (1925) studied certain
physiological distinctions between fresh-water and marine organisms.
He found that marine organisms show a much greater toleration for
fresh water than fresh-water organisms do for sea water. His observa-
tions tend to support the long-accepted belief that organisms migrate
from the ocean into fresh water, rather than in the opposite direction.
In an interesting and suggestive study Marshall and Smith (1930) have
attempted to correlate the composition of the body fluids of marine and
fresh-water fishes with renal function and to trace the evolution and
migration of these vertebrates on the basis of changes in the structure
and activity of the excretory organs. Pantin (1931) has studied the
trie-lad turbellarian (inmht uli'cc. which occurs in the estuaries of small
streams, and found that these anelomate worms withstand both fresh
and salt water, and that in nature they may be exposed to either extreme
DILUTION OF SEA WATI-.R AXU CERCARLE 245
for several hours. In tap water they double their volume and lose 25
per cent of their salt content in an hour. The presence of calcium re-
duces the rate of swelling and the loss of salts, presumably by reducing
permeability. The significance of calcium and its relation to the prob-
lem of the migration of animals into fresh water was discussed.
The difficulties of migration from one habitat to the other would be
greater in the case of parasitic species like the digenetic trematodes than
in free-living forms. Where two or more hosts are involved, and where
the transfer to the next host is effected by very delicate, short-lived,
larval stages, the initial obstacles to migration are augmented by the
difficulties inherent in the completion of the life cycle. For such a
parasite to change from one location to another, either both primary and
secondary hosts must have made the same migration simultaneously, or
the parasite must have changed to new hosts as the migration progressed.
Furthermore, and probably of greatest importance, the free-living,
aquatic, larval stages must be able to withstand the changed environ-
mental conditions and remain infective.
The migration of primary and secondary hosts has not been ex-
tensive in recent times at least. The groups of mollusks, fishes, reptiles,
and mammals are clearly separated into marine and fresh-water species
and this distinction has persisted with but very little change since earlier
geological time. The shells of mollusks and skeletons of vertebrates
afford suitable material for fossil formation and the geological history
of several of these groups is known. According to Zittel (1913) "By
means of analogy with recent species we are able in most cases readily to
determine whether fossil forms pertain to land, fresh, brackish, or salt-
water species." He stated that " not until the boundary between the Jura
and Cretaceous is reached do we find any traces of fresh-water snails.
... In the Wealden, and Cretaceous generally, both land and fresh-
water gastropods are quite abundant ; they become highly developed and
widely distributed during the Tertiary, attaining, in fact, a differentia-
tion nearly equal to that exhibited by the corresponding recent forms."
Since the vertebrate hosts are more active, wider ranging, and longer
lived than the molluscan hosts, it would appear probable that if migration
is to be considered as the explanation of present distribution, the verte-
brate hosts were the principal migrants and that they were primarily
responsible for change of habitat. The paleontology of the turtles
(Hay, 1908; Williston, 1914) indicates that the marine turtles and the
soft-shelled, fresh- water turtles have been separate, independent groups
since the Mesozoic era. Looss (1902) described several genera of
pronocephalid trematodes from marine turtles; Stunkard (1930) and
Mackin (1930) have described members of the same family from the
17
246 H. \V. STUNKARD AXU C RUTH SHAW
fresh-water turtles, Ainydu and Pscitdcinys. The discovery of related
parasites in hosts that have been separated since the Mesozoic would
suggest that migration of hosts is not to be accepted as an explanation of
these cases at least.
There an- also serious objections to the explanation involving trans-
fer to new hosts. While host parasite specificity is not so limited as
was formerly believed, and it is well known that many parasites may
infe-t several host species, as a rule the possible host species are closely
related. For this hypothesis it is essential also that both old and new
hosts live in the same habitat, since otherwise they would never en-
counter the infective larval stages of the parasite. Consequently, if
separation into marine and fresh-water species was effected by the adop-
tion of new hosts, the transfer could occur only in those regions where
fresh and salt-water habitats overlap, namely at the mouths of rivers.
In the transitional zone of brackish water, with the recurrent increase
and decrease of ihr salt content and pi I of the water with the rise and
fall of the tide, transfer to new hosts may have caused divergence into
definitely marine and fresh-water species.
It is thus possible that both migration of hosts and transfer to new
hosts or a combination of the two methods may have been operative in
producing present distribution of related species. It may be that the
distribution of existing groups of digenetic trematodes is correlated with
the origin of these groups and this point should be considered in any
treatment of the problem. The present complicated developmental
cycles could not have been the original or primitive life histories of these
species. It has long been recognized that parasites have been derived
from free-living progenitors. Competent investigators agree that the
trematodes and cestocles have a turbellarian ancestry. The subject was
discussed by Meixner (1926} with the following summary, " F.s ergcbcn
>ich drei Schliisse :
I. Dass die Differenzierung der Trematoden und Cestoden mit
clem Auftreten der \Virbeltiere eng vcrkniipft ist.
" II. Dass die /urn Parasitismus auf Kvertcbratcn ubergegangenen
Vorfahren der I )igenea und ('esloden entsprechend der heutigen
Beschrankung der primiiren Larvcn hen-its auf verschiedcne Wirtstier-
klassen spezialisierl waren.
" III. Trematoden und Cestoden sind zwei infolge ihres Para-
sitierens auf \Yirbeltiercn insbcsonders hinsichtlich des Integumentes der
Reifestadien abgeandertc Anhangsgruppen der Rhabdocoela."
Concerning the origin of these groups Reisinger (1928) stated.
" Rczcichnend fiir die Amera ist die in vielen (Iruppen vorherrschende
Neigung zu parasitarer Lebensweise. vielleicht in . \usniitzung einer
DILUTION OF SKA WATKR AXD CKKCARI^E 247
besondm'n. dan ganzen Unterstamm eigenen, stoffwechselphysio-
logischen [Constitution, die den einzelnen (iruppen den Ubergang zu
intramolekularer Atmung (Glykogenabbau) besonders erleichterte.
Die Urheimat der Amera ist zweifellos das Mecr ; der Ubergang zu
terrikoler nnd parasitischer Lebensweisc mag sowohl von dort ans wie
aucb vom Siisswasscr erfolgt sein nnd erfolgen."
Bresslau and Reisinger (1928) concluded that, " Unter den Rhab-
docoelen verdienen die Familien der (Iraffilliden nnd Anoplodiiden
besonderes Interesse, insofern als von ilinen ans vermntlich die Ent-
wickelung der Trematoden ibren Ausgang genommen hat. Nach ihrer
ganzen Organization sind die Monogenea wahrscheinlich von Graffilliden,
die Digenea von Anoplodiiden oder anoplodiidenahnlichen Kalyptor-
hynchiern (Rhabdocoela) herznleiten. Gut stimmt damit iiberein, dass
gerade diese Familien das Hauptkontingent an Parasiten nnter den
Stirudelwurmer stellen."
Consideration of this subject raises one of the most difficult prob-
lems in biology, the origin of intermediate hosts and the digenetic life
cycle. The original ancestors of the digenetic trematodes must have
become parasites of aquatic animals and the evidence indicates that
mollusks were the original hosts. The type of reproduction in these
mollusks is problematical. It is well established that parasitism in-
creases reproductive activity, that it leads to new and accessory methods
of reproduction, and that asexual multiplication is frequently inter-
polated between the sexual phases. There may have been a sexually
mature, free-living stage after asexual multiplication was developed in
the invertebrate host. The appearance of the vertebrate host may be
correlated with the evolution of vertebrates and their use of mollusks
as food.
Presumably the adoption of the parasitic habit occurred at an ex-
tremely remote period and the evolution of parasitic life histories and
accompanying transformation of the parasites have proceeded hand in
hand with the evolution of their hosts. The parallel evolution of hosts
and parasites has been demonstrated by many authors. The presence
of related species in both marine and fresh-water hosts may be explained
by assuming that the primitive hosts harbored the ancestors of present
species, and that the hosts have subsequently separated and differentiated
into marine and fresh-water species. Such an explanation would imply
that descendants of the original hosts have carried their parasites with
them since the separation and, as a result of the ensuing migration and
modification, the primary hosts, secondary hosts, and parasitic species
have evolved together. It may be contended that this explanation
merely pushes the problem further back in the time scale, renders it
^^ \ \
248 H. W. STUNKARD AND C. RUTH SHAW
more difficult of analysis, and less susceptible of experimental treatment.
\Yhile to a degree this criticism is valid, the postulate may nevertheless
be correct, and there are, moreover, experimental means of investigating
the question.
Studies of much modified parasitic species and interpretation of
their life histories are greatly facilitated if the life cycle contains free-
living larval stages. These stages presumably correspond to ancestral
ones, since it is generally true that embryonic and early developmental
stages are very conservative and tend t<> remain unchanged regardless of
modifications which may occur in the later development of the animal
concerned. Since digenetic trematodes have such free-living larval
stages, and since these larvse are the infective agents, providing for the
transfer from one host to another, — an experimental study is possible.
Knowledge concerning the effects of environmental changes on these
larvse may have significant value in the interpretation of life cycles and
distribution. Since the trematodes have two free-living larval stages
in the life cycle, data should be obtained for both the miracidial and
cercarial stages. It is often difficult if not impossible to secure miracidia
in sufficient numbers for such experiments, while cercariae can usually be
obtained in abundance. It therefore seemed pertinent to make a study
of the effect of the dilution of sea water on the activity and longevity of
marine cercarise. A corresponding study, already started, on the effects
of diluted sea water on fresh-water cercariae will give data, which, cor-
related with those from the present investigation, may aid materially in
explaining present distribution of related species in marine and fresh-
water hosts.
No matter whether the present distribution is explained through
migration of original hosts or transfer to new hosts, the essential factor
involved is the ability of the free-swimming larva? to live and remain
infective in the new environment. The ability of these larval stages to
function in increasing or decreasing salinity indicates the direction of
migration and the original home of the original trematode species.
Consequently, the experiments reported in the present paper were
undertaken.
MATI.RIAL AND MKTHODS
All of the cercari.e used in the investigation were obtained trom
mollusks of the Woods Hole region, and the experiments were done at
the Marine Biological Laboratory during the summer of 1030. Data
are given in the tables for the following six species: (1) the cercaria of
Cryptocotylc linf/ua from Littorina littorca. (2) Ccrcariccnni lintoni from
\'assa obsolcta. ('3) Cercaria quissctcnsis from N. obsolcta, (4) C. vari-
{ilaiidis from .V. obsolcta. (5) C. parricaudata from /,. littorca, and (6)
C. sen si 'f 'era from Urosalpinx cinercns.
DILUTION OF SEA WATER AND CERCARI^ 249
The snails were isolated in small dishes of sea water to determine
those from which cercariae were emerging. Several of those infected
by one and the same species were then placed in a small dish for 10 to
12 hours in order to secure large numbers of recently emerged cercaria^.
At the end of this period the snails were removed and the cercarise trans-
ferred to small dishes of sea water, usually 20 to 50 in each dish. The
sea water was removed from these dishes and replaced by sea water to
which various amounts of tap water had been added. Solutions were
made up as follows: (I) undiluted sea water; (II) ^4 sea water, *4 tap
water; (III) y2 sea water, l/2 tap water; (IV) l/4 sea water, y\ tap
water; (V) l/% sea water, ?£ tap water; (VI) tap water. The dishes
were covered to prevent evaporation and kept at the temperature of the
laboratory. In each experiment all of the larvae were subjected to
identical conditions except for the different amounts of tap water in the
solutions. The only variable factor, therefore, was the amount of tap
water and the results show the effects of increasing dilutions of sea
water. Observations were made with a binocular microscope at appro-
priate intervals and the condition and activity of the larvae noted.
Ccrcaria of Cryptocotyle lingua
An abstract of this experiment was reported (Stunkard, 1930r).
At first the larvae are very active and all swim vigorously by rapid
lashing of their tails, holding the body motionless in a curved position.
Swimming movements cause the cercariae to rise toward the surface of
the water and when swimming is temporarily suspended the larvae
slowly sink. They are positively phototropic and accumulate at the light
side of the dish. As the vitality of the larvae diminishes they become
progressively weaker and are unable to rise from the bottom. This is
due primarily to exhaustion of the tail muscles. The larvae then extend
and retract their bodies and tails, although since there is no functional
acetabulum they can make little forward progress. In the solutions
which contain 50 per cent or more of tap water the tails soon begin to
swell and lose their motility, and later the body swells. Naturally the
swelling is more rapid and greater in the more dilute solutions. Ap-
parently swelling is inhibited so long as the tissues are alive and active.
The swelling causes cytolysis of the tails and they soon drop off. The
tails are frequently lost after a few hours in all of the solutions. In
the tables the following notations are used :
Swimming vigorously -| — \--\ — \-
Swimming seldom and feebly -\ — ( — \-
Contracting vigorously -| — \-
Contracting feebly -f-
Dead —
250
H. W. STUXKAKD AND C. RUTH SHAW
The experiments \\rre repeated four times and the results arc in
substantial agreement. The following protocol is representative.
_ -1
.
.
H =
M ++ +1 +1
- o f^i -co — —
— x; c\ ac so i^
2 +
^1 C OC
•*-, — sc
g
3C CN l/V -t »- ^C —
O O ir, _
+
+ + +
vC
+
«3 rt rt
+ ++ I
O >O <M CN
•O
O O
O rs
O
q
<~o
o
o
0
DILUTION OF SEA WATKR AXD CKRCARI.V.
251
•"•"* »— •
i_
> ?
+ +
+ + 1 +
t^. 10 t^ o
<T5 CTs ^H
1 +1 +1
O m 'O 10 10
+ i i
Ol 00 O
fy^ *^4
> 0
+ + + +
++ 1 ++ 1 ++ 1
CO t^* ^O IO (O IO IO CO *O
ro i-i 10 O
i ,
r o 10 o
"~. O t— •
t+ :
+ + o
o
'O
-i — [- -f- +
H — h H — h +
_)_
+ + + i +
++ 1 ++ 1 + 1
ro *— i fO *^
+ 1 1
'O fO rO
IO »O
i— i ON
— ! O
++ +
+ + +^
+ o
+ + + i +
1^ OO Ol Ol PO
+ 1 ++ 1 ++ 1
^^ *O •»-"< ^^ GO ^O ^O CO
*-* -t1 O) ro CN
+ + 1 + 1
IO VO OO IO ~t<
OJ re O
01
++ +
+ t
»
+ + + +
^^. rv^ f*^ QQ
++ I ++ 1 + 1
•^ 1^-1 r^
+ 1 +1
01 0)
J
C^ ""-
8
CA
fi o
o ^^<
'£ o . o
b/iO
9. ™o
ex; <! c^
T— * ^-*
10 MO
— 3 O
-C < ^
"o »^ ~ •—
c/3 Z<1H
H. W. STUXKARD AXD C. RUTH SHAW
Analysis of the data shows that Solutions II, III, and IV have only
slightly harmful effects as the sea water is diluted. In solutions con-
taining 50 per cent or more of sea water the effect is not significant and,
in twn of the experiments, after 12 hours in Solution No. II the cercarije
were more active and vigorous than those in undiluted sea water. In
Solution X<>. Y. which contained l/$ sea water, the cercarue were all on
the bottom of the container at the end of 2 hours. They had begun to
swell noticeably, some had lost their tails, and those whose tails beat
rapidly were unable to rise in the water. It is probable that the larvae
are not infective in this concentration for more than a few minutes.
The range between ]As and *4 sea water appears to be the critical zone
where the dilution of the sea water exerts a markedly harmful effect
on the physiological processes of the larvae. Freshly emerged cercarue
manifest normal swimming movements for only a few minutes when
placed in tap water; at the end of 20 minutes all had settled to the bot-
tom, in 30 minutes about l/2 of them showed no sign of life and the
others soon succumbed.
Ccrcariccnin Union/ Miller and Northrup, 1^26
Solutions were made up as in the previous experiment and the same
procedure was followed. Since these larva have no tails they can not
swim and their activity is restricted to creeping movements as described
by Miller and Xorthup (1926). The experiment was repeated seven
times using 20 recently emerged cercarise in each dish. In two of the
tests distilled water was used instead of tap water and the larva,- lived
as long, and in certain of the dishes slightly longer than in those con-
taining the same amount of sea water diluted with tap water. The dif-
ferences were not great and probably are not significant. The results
are similar for all experiments and the following protocol, given on
p. 253, is typical.
These results are similar to those obtained for the cercnrife of C.
liiu/na. The larvae show very little normal activity after 15 minutes in
tap water and it is apparent that they are not infective in this medium.
Two larva? encysted on the bottom of the dish but it is apparent that
encystment in the water is not a usual or normal stage in the life history
of the species.
Ccrcaria quissclcnsis Miller and Xorthnp, 1926
The experiments were conducted as previously described and re-
peated eight times, using 20 cercarise in each dish. In swimming, the
DILUTION OF SEA WATER AND CERCARI^E
253
> 04
o
01
u u u 11
•
000
-i. ^
t-c U
•* 01
+ + + , 0 0
+ + + i £ i ++ + + + i
1 1
C C OO 01 «-> i— OJ CO , i- ro i-t Ov
o o
01 01
u | u | u |
| |
0 0 0
^^ ^ "*
u u u
UM t^
i— i 01
+ + + + +0+0+0
O O
+ + + + ++ ++ ++
+ + ++ 1
C O O O iO>O 0400 •>* O
^— • OS ^H CM i^
01 01 01 01 *-H ~* >-i
^— ( i-H
u. | i-| u |
| |
0 0 0
1— 1— i
^— 01
+ + + + +0+0+0
O O
+ + + + ++ ++ ++
+ + ++ 1
OOO C O'^f'OiO'^sO
01 GO i-H 01 1^
Ol O4 Ol Ol *— i "—1 -H
I 1 1
U 1 L- 1 1
1
0 0
i— i O
u _l_ *" u
J^
i— 1 Ol
O
+ + + + ++ ++ ++
+ + 1
O O O O sO ^f Tf sO ""3 LO
04 OO O
Ol Ol O4 O4 *-i *-H »H
>— i Ol
1-1 1-1 1-
0 ' 0 ' 0
O
u i_
Ol
+ + + + +0+0 +
+ + + + ++ ++ + 1
+ 1 +1
^D ^^ ^5 ^D ^* **O *O LO '~~l O^
fO 1^- ^ — ! Os
01 01 04 Ol *-H -^H >-i
1 £
— 01 so 01 -f O OO
O O
rd 5
»— i O4 ro -4"
so r^
u *r<
t/i
~ U r"
W- •-
= - S
a Is
254 II. \V. STUXKAKD AND C. RUTH SHAW
body assumes a spherical form and the tail lashes vigorously. Records
were taken every hour and the following protocol, given on pp. 255 and
256. tabulates the results of one experiment.
In this species the cercaria? lose motility after 15 to 30 minutes in
tap water. The tails swell in all the solutions containing 50 per cent or
mt in- ot" tap water and soon become detached. There is a tendency for
tin cercaria,' to encyst after 24 hours in the more concentrated solutions
and such larvae removed from their cysts at the end of 72 hours were
alive and active. There were only slight differences between the cer-
cariai placed in Solutions I and II. During the first half of the experi-
ment the larvre in Solutions III and IV appeared to be affected more
than those in Solutions I and II, but in three of the tests they lived
lunger than those in undiluted or 75 per cent sea water.
Cercaria nir'njlanilis Miller and Nortlntp, 1926
This species is very rare. Miller and Northup found only 3 in-
fested snails among 8,875 individuals of Nassa examined, and we found
oiilv 2 infected snails. The structure of the cercaria indicates that it is
./
the larva of one of the blood flukes, and it swims in active spurts. Only
two experiments were made, but the results, given in the protocol on
p. 257. indicate that the larva; arc short-lived and very delicate. The
procedure was the same as that previously employed.
As the cercariae lose motility they become distorted, the furcae coil
up and it is sometimes difficult to determine whether or not they are
dead. Soon, however, they turn dark-colored and later they tend to
float.
parvicaudata ii.sp.
Two experiments were made with these larvae. The procedure was
the same as that previously employed, although the observations were not
continued until the death of the ccrcari.-e. One of the protocols is given
on ]>. 25cS and the other is in essential agreement.
The cercariae in Solution VI (tap water) were all dead and their
bodies much swollen at the end. of one hour. At this time all of those
in Solutions I, II, and III were swimming intermittently, while those in
Solutions IV and Y were unable to leave the bottom. They were all
lying on the dorsal side, bodies bent as in swimming, with the tails mov-
ing. After four or five hours they seemed to be more active although
they were unable to leave the bottom of the container. At the end of
25 hours, although they were unable to swim, the larvae in Solution II
were more active than those in sea water, and those in 50 per cent sea
water were more active than those in Solution II.
DILUTION OF SKA \V.\TKR AND CKRCARI/E
255
o
01
O
01
0 +
O
Ol
to
0)
u
+ o $ o
.-I- + -f 2 ++2 +|
OO O10000 CNOO>O -tO
> 0
_ , 01
o
Ol
o
O)
u+ Eg
O 4>
OO Ol
tfl
vs
0)
HI es
o
Ol
o
01
O
01
O
Ol
cn
OO 01
+ O »
o
01
+
O
01
O
01
O
Ol
o
Ol
!-, 01
o ju
OO 01
oo O 01
+
+
O
04
o
01
O
01
i_ en
o _o
<-j —
«3 ~
CJ ^J
c
o
O
c«
u
"o
6
o
K
H
-f
01
256
H. W. STUNKAKD AXD C RUTH SHAW
o
CM
+ + I
oo ^o ^^
s '
u
+ +
1_
+ o
+ +
+ 5
+ + I
+
en
+ <-> I
T- — 1 OC
O
+ o
en
c/l
U
U3
!fl
C
0
c- -
O
+
+
+ I
CM X
~
<M
+ o+ I
t-- CN O >O
• **
-)-!/)
+ o
u tn
O
+
CN CN C
</)
tfl
tmt
in
OC
O
CO
-t
O
I-
£
_O
*-•
_3
S
&
"c "I
£ '£•
DILUTION OF SEA WATER AND CKRCARIJE
257
, ,
+ 1+1 1
> CN
0! CO O -t O
W-H ^~* C**'!
> o
O i —
CN
ft u +,
IO CN OO O Tj<
1
O
T— ( 1O i — i ^-^
CN
> o
u
i— i CN
-.1— — 1-. o
+ ++ +
s '
+ + ++ ++
+ o
+ + ++ ++ 1
+ + 1 + I
2 "" ^ ^ ^ ^ *°
^ *o o 01 oo
1- '—
O O
~ 0
" 01
+ +
i
+ + + +
+ o
+ ++ + +
+ + 1 + 1
o o •— i "~- >o
10 ^o oj r^ oo
01 — i .-I
'~H '~i
u
^H ^J
O
+ + + +
+ + i i
O O 1^- ^O
t^. 10 oo o
C^l 01 ^H
CN
^^
+ + + +
+ + 1 1
c o •* -o
-* CN •* O
CN 0) •^
— ' CN
tt
rt §
^H CN O
01 -t-
*T*
= ° e
**- "*
O Qj
- 6 S
<8 fc £
258
II. \V. STUXKARD AXD C. RUTH SHAW
o c
++ I ++ I
CC ""> 1^ 1C 1^ OO
~: -t f r-i
OC
O O
\r,
000
•O *>O ^O
o
+
r
~
ro
o
+ 1
I =
5 S
.E
"o u
c^
DILUTION OF SKA WATER AND CKRCARI/T*: 259
Ccrcana scnslfcra n.sf>.
Four experiments were performed with these larvae, and the results
of one experiment arc tabulated in the- following protocol, given on
p. 260.
There is a pronounced tendency for these cercariae to encyst when
subjected to unfavorable conditions. The process of encystment is
rapid and the cysts are either free or attached to the bottom of the con-
tainer. Presumably this phenomenon is normal and significant for life
history studies of the species.
Cercarise emerge usually at night or in the early morning and the
majority soon encyst. They tend to adhere to any object they touch and
numbers stick to the inside of a pipette used to transfer them from one
solution to another. After attachment they soon encyst.
DISCUSSION
In an investigation of this character, it is desirable to study as many
species as possible and representatives of different taxonomic groups.
Unfortunately, information concerning the marine larval trematodes of
North America is very meager. Only a random sample of the species
has been described. The literature dealing with these larvae was re-
viewed by Miller and Northup (1926), who described five species from
Nassa obsolcta at Woods Hole, Massachusetts. It is significant that
only one of the five had previously been reported. Since these were
almost the only larval trematodes described from the Woods Hole region,
an attempt was made to secure them for the present study. All of the
five species described by Miller and Northup were found and three of
them in sufficient numbers for the experiments. Of the other three
species studied, one was shown by Stunkard (1930a) to be the larva of
Cryptocotyle lingua, while the two remaining species are new to science
and are described in a later section of this paper.
For these experiments it is essential that cercariae be available in
large numbers. Since cercariae secured by crushing parasitized snails
are immature and not infective (Stunkard, 1930, 1930&), such larvae
do not constitute suitable material, and results obtained from them are
probably not significant. Consequently, only normally emerged cercarire
were used. Since several hours are required for the emergence of
sufficient numbers, some of the cercariae had been swimming for ten to
twelve hours when the experiments were started. This factor un-
doubtedly accounts for much of the variation shown in the results.
Presumably the most recently emerged cercaria lived the longest.
It is apparent in all species studied that tap water exerts an im-
mediate and harmful effect. None of the ccrcarise showed normal
260
H. W. STUNKAKD AXD C. RUTH SHAW
> *
1
O
+ 1 1
'^ s
r^i oo
j_ "O _t_ T! "O T
ft/ 4) ft) &
— —
— |— t/l — |— CO (O 03
s/> y5
- -1
^* ^ ?•% ^\
-4- <J -+• -4- u u o
-j- 4> ++ «U | + £ | + 5 |
•o' C
+ S | a) |
oc**" -tfN-tr^i vcxcc; CNO-^
c^l O -t vO OO
01
+ _!_ ~ i *O i "S
~ r: 0) ~T 0)
*J "^
o> <u
^ HI
4- + > 4- u5 + 'S
uj ^n
+ + ++ f + u
-j- O O
O1 OC*— "", fj '•^ f^^ —
-t- -t- — ir, -r
+ + + +-s
_l_-o -a
1^ <l> 1^
+ + + +*
— (- C/3 C/)
— 'j;
+ 1 1 1 ^
-4— -4— —1— CJ
i C
+ M ?
+ + + + s
i QJ g^ i
U -H ^ ^ "
OX ^H O
•r- 1
T ~r rt i £ i 4)
u £
— ~
+ +" + * + 1"
u5 "S
c =
+ H — h H — h41 +**
-f ~-. — ~l — -H O OO
C? u
:s i s i
• ^C f^ OC O
« E
^H rv^ sO 'O
« £ ^
§ ^ -S
'£ O D
1 6 B
(/5 y. H
DILUTION OF SEA WATER AND CERCARIA 261
activity for more than a few minutes and most of them died within an
hour. The bodies and tails became swollen, the • tissues underwent
cvtolysis with the absorption of water, and death followed shortly.
Presumably there was a diffusion of salt from the organisms as water
was absorbed and the loss of salt would augment and hasten the
deleterious effects produced by the imbibition of water. Obviously these
cercaria? can be infective for only a very brief period in tap water and
it is doubtful whether they could complete their life cycle in fresh water,
even if suitable hosts were available.
Larvae placed in Solution V, containing seven-eighths tap water and
one-eighth sea water, were active for considerable periods of time and
some of them were able to perform swimming movements for one to
four hours. Certain of them, e.g., Cercaria parvicaudata, appear to be
deleteriously affected after a short time in this concentration, and later
they partially recover. They may continue to live for several hours,
although the earlier ill effects are not entirely remedied and it is doubtful
whether such larva? would be infective. They are not infective in the
case of C. lingua and since the life histories of the other species are un-
known, experimental test is impossible.
The experiments show a marked difference in the activity and
longevity of cercariae in Solutions V and IV. Whereas a solution con-
taining one-eighth sea water is definitely harmful, cercaria? manifest
little in the way of ill effects in solutions containing 25 per cent sea
water. In one-fourth sea water the larvae live almost as long as in
greater concentrations, although they are usually less active after the
first few hours. The sluggishness may be due to the increased water
content. Cercariae may succumb somewhat more quickly in one-fourth
sea water than in more concentrated solutions, but they are normal in
appearance and activity for sufficiently long periods of time to indicate
that they may be infective and able to function normally in continuing
the life cycle.
Considerable interest attaches to the observation that larva? are active
and apparently normal for almost if not quite as long in solutions con-
taining 50 per cent or more of sea water as they are in undiluted sea
water. In certain experiments, cercaria? actually lived longer in one-
half than in undiluted sea water, although they were not normally active
and probably not infective for longer periods than larva? in sea water.
The ability of cercaria? to withstand dilution of sea water is roughly
proportional to the dilution which occurs in the larger bays. Cowles
(1930) reported that, "The salinity of Chesapeake Bay, like that of
other long bays and estuaries, gradually decreases, with very few ex-
ceptions, from the mouth to the head ; and the bay is known as a
18
262 H. \V. STUXKARD AXI) C. RUTH SHAW
brackish body of water, although the failure as a rule, of the fresh
waters from the land and the saline waters of the sea to mix completely,
and the variation in the volume of fresh and salt water entering the bay.
result in different degrees of brackishness. The surface data at the
mouth of the bay show a variation in salinity from about 19 to 30 grams
r liter, while near Baltimore there is a variation from about 3 to 11
-nuns per liter." . . . "The bottom salinities recorded on our cruises
for the mouth of tin- bay varied from about 26 to a little over 3J. while
iu the region of Baltimore they varied from about 6 to 17."
It appears more than probable that such transitional zones, extending
sometimes for a distance of a hundred miles or more, with ocean water
at one end and fresh water at the other, provide the ecological settings
in which species become transformed physiologically and structurally
from marine to fresh-water organisms, and vice versa. Due to the well-
known and constantly appearing variations which occur among animals,
certain fresh-water species may have become adapted to life in brackish
and eventually sea water, while similarly, marine species may have
eiuc-red fre^h-water habitats.
If the larvae of parasitic species are able to survive in a new and dif-
ferent environment long enough to find and intect a suitable host, either
a former host that has migrated into the location, or a new host species,
the life cycle may be completed. The long list of snails which serve as
intermediate hosts of I'usciola hcpatica, under different conditions in
various parts of the world, demonstrates the extent to which that species
has acquired new hosts and the work of Cort (1918) illustrates the
ability of other trematode larva* to successfully attack new hosts. In the
case of a marine or brackish-water species entering fresh water, the es-
sential factor is the ability of the free-swimming larva? to withstand the
hypotonic medium until infection is accomplished, since in the body of
the host a medium of higher salt content and tonicity is encountered.
The present experiments record the abilitv of six marine cercari?e
to withstand dilution of sea water and show that these larva' manifest
normal activity for considerable periods of time in solutions containing
only one-eighth to one-fourth M-.-I water. The observations indicate thai
these cercarise are able to complete their life c\des in brackish water and
denote the extent to which marine species may migrate into brackish and
fresh water. The paper thus contributes toward the solution ot the
problem of the distribution of related species in marine and fresh-water
habitats.
DILUTION OF SEA WATKR AMD CERCARLE
PLATK I
26.3
FIG. 1. C. scnsifcra, ventral view.
FIG. 2. C. scnsifcra, excretory system.
FIG. 3. C. scnsifcra, ventral view, showing distribution of gland cells.
FIG. 4. C. scnsifcra, cross section, showing three types of gland cells.
FIG. 5. C. parvicaudata, ventral view.
264 II. \V. STUNKARD AXD C. RUTH SHAW
Ccrcaria l^arricaudata ;?..?/>.
(Fig. 5)
This species occurs in about one per cent of the specimens of
Littorina littorca examined at \\"oods Hole. The cercariae are produced
in sporocysts which occupy the lymph spaces of the snail. The gonad
is the principal seat of infestation and frequently this organ is entirely
destroyed.
The cercaria? are small, with oval to pyriform bodies and short tails.
The length of the body varies from 0.14 to 0.36 mm. in contracted and
extended condition. The tail is very active; it may be contracted to a
length of 0.06 mm. or extended until it exceeds the body in length.
Ordinarily when the worm is attached or creeping, the tail is contracted
and manifests a nervous, twitchy motion. In swimming the body is
contracted into a short, wide form, bent ventrally and the tail is extended
and lashes violently.
The body is covered with minute spines and the oral sucker bears
a stylet 0.015 mm. long by 0.0032 mm. wide. There is a small thicken-
ing on the stylet near its tip. The acetabulum is situated near the middle
of the ventral surface and measures from 0.03 to 0.05 mm. in diameter.
The cercaria' encyst readily. A portion of a dissected snail was left
for six hours in a watch glass and ten cercaria.' had encysted among the
sporocysts. Other cysts were found in the tissue of the snail which had
been fixed and later sectioned. The cysts measure approximately 0.17
mm. in diameter. There are gland cells of two types distributed
throughout the parenchyma of the body; one is filled with refractive,
spherical granules, the other is slightly opaque and contains very fine
granules. On either side of the mouth there are openings of ducts which
pass backward and appear to communicate with other glandular cells
situated in the preacetabular region, although the connections of these
ducts were not determined with certainty. Presumably they are the
ducts of penetration or salivary glands.
The oral sucker is spherical to oval. 0.035 to 0.06 mm. in diameter.
It is followed almost immediately by a small pharynx and the esophagus
extends about one-half of the distance to the acetabulum. The digestive
ceca terminate blindly near the level of the caudal margin of the
acetabulum.
The excretory vesicle is Y-shaped, with a long stem and short
branches. Its wall contains large, deeply-staining cells. Four flame
cells have been definitely located and these are shown in the figure.
< Mhers were observed, but their connections were not traced. '1 he re-
productive organs are represented by a mass of cells which arc dorsal
and anterior to the acetabulum.
DILUTION OF SKA WATER AND CERCAR1/E 265
This species belongs in the large and heterogeneous group of
Xiphidiocercarise, but further attempts to relate it must await a more
complete knowledge of its morphology or information concerning its
adult form.
Ccrcaria scnsifcra n.sp.
(Figs. 1-4)
This species has been found onlv in the ovster drill, Urosalf>ln.\'
cincrciis, and il was present in fourteen out of 594 specimens examined
during the summer of 1930. Six infestations were found in two hun-
dred and six snails collected at Woods Hole during the first week of
April, 1931. The parasites infest the interlobular areas of both the
reproductive and digestive glands. In an uninfected snail the visceral
mass is plump, the liver is yellow and the gonad is cream-colored,
whereas in a parasitized snail the organs are shrunken, the gonad may
be destroyed, and the body is lighter in color.
The cercarise (Fig. 1) are large and clearly visible to the unaided
eye. The body is oval in shape, more or less elongated and narrower
posteriorly, flattened dorsoventrally, and widest in the preacetabular
region. It is truncated posteriorly and the attachment of the tail is
terminal. The cercarise vary considerably in size and manifest much
elongation and contraction in locomotion. They are not active swim-
mers and tend to remain near the bottom of the water. After a time
swimming movements alternate with creeping ones. In swimming, the
tail is elongated ; it does not lash about, but the cercaria moves by nn-
dulatory movements of the body and tail. The chief propulsive force
comes from the anterior half of the body. It slowly bends ventrally and
then snaps backward, pulling the larva forward. The movement is con-
tinued through the posterior part of the body and tail producing the
sinuous motion of the larva.
With the exception of the anterior end, the body is covered with a
thick granular cuticula which bears large, closely set spines. Those in
the anterior row are considerably larger than the others. This row is
interrupted in the midventral region. There are 44-48 spines on the
dorsal side and 10 on each side ventrally. These spines measure 0.005-
0.006 mm. in length. This region of the body is sometimes contracted
to produce a distinct collar-like effect. There are about forty-five an-
nular rows of spines in the preacetabular region and about 130 to 135
such rows on the body. The spines in successive rows alternate with
each other and those around the acetabulum are arranged in concentric
rings. The cuticula of the tail is thin and smooth.
The larvae are bottom forms, and attach readily to any available
266 H. \V. STUXKARD AXD C. RUTH SHAW
surface. \Yhen picked up in a pipette they frequently adhere to the
inside of the tube and can be dislodged only with great difficulty. They
may become attached either by the suckers or by the tip of the tail and
after attachment they soon encyst. If placed in solutions that are ir-
ritating, e.g., too strong concentrations of vital dyes, they encyst almost
immediately. The cyst consists of two layers, a thick, opaque, external
covering and a thin, transparent, very tough, inner membranous layer.
Neither of the cyst walls is readily stainable by ordinary dyes. The
cyst is oval, flattened on the side of attachment, and measures from 0.2
to 0.23 mm. in width by 0.23 to 0.27 mm. in length. The tail is always
detached in encystment and may remain attached for a time to the sur-
face of the cyst. Normally the worm fills the cyst completely. The
outer cyst wall is easily removed by rolling a cyst between a slide and
cover glass, but it is difficult to get the worm out of the inner mem1-
branous covering without injury.
Living cercarire may extend to a length of 0.9 mm. and contract until
the length is no greater than the breadth. The tail also is capable of
much extension and contraction ; it may be very much shortened or
elongated to almost the length of the body. In the latter condition it is
Blender with an expanded, cup-shaped portion at the end. The caudal
tip is usually introverted in a characteristic manner (Fig. 1). although
the invaginated portion may be protruded and apparently bears a sticky
substance, by means of which the cercariae adhere to objects. Fixed
and stained sperintens measure from 0.21 to 0.47 mm. in length and 0.14
to 0.26 mm. in width. In such specimens the tails vary from 0.12 to
0.26 mm. in length. The acetabulum is situated slightly behind the
middle of the body. In living specimens it measures from 0.08 to 0.1
mm. in length and from 0.1 to 0.115 mm. in width, while in fixed and
stained specimens it measures from 0.68 to 0.76 mm. in diameter.
The anterior end of the body is covered by a smooth, thin, unarmed
cuticula which extends backward as far as the caudal margin of the oral
sucker. This region bears a number of papilhe, arranged irregularly
in two or three rows, and each papilla terminates in a bristle. Similar
structures occur around the margin of the acetabulum and presumably
they function in a sensory manner. This idea is expressed in the spe-
cific name, scnsifcra. The caudal one-fourth or one-fifth of the anterior
unarmed area forms a zone which frequently is marked by small longi-
tudinal furrows.
The body is filled with gland cells of several types. In the region
between the pharynx and acetabulum there arc twelve large unicellular
glands. These cells (Figs. 3, 4. r//r) have very small secretory granules
and open to the surface through twelve pores situated at the anterior tip
DILUTION OF SKA WATER AND CERCARI/E 267
of the body above the oral sucker. They seem to correspond to pene-
tration glands of other cercarise and indicate that the larvae at some later
stage bore into the tissues of an intermediate host. The cortical layer
of the .parenchyma contains numerous dermal glands. In addition, the
dorsal half of the body contains numerous gland cells (Figs. 3, 4, gla),
the cytoplasm of which is filled with bacilliform granules or rods.
These cells do not stain with neutral red and in sections counterstained
with erythrosin the secretory products appear yellowish. The ventral
portion of the body is largely filled with gland cells (Figs. 3, 4, gib)
whose cytoplasm contains large spherical secretory granules. The cell
contents stain intensely with erythrosin. In the anterior half of the
body these cells appear to be arranged in four longitudinal fields, sepa-
rated by the large ventrolateral nerve trunks and the esophagus.
Attempts were made to study the cercariae in solutions of various
vital dyes. With neutral red the spine-covered portion of the cuticula
and the cells which secrete it quickly take up the stain and this red or
pink layer at the surface of the body masks the action of the stain inside.
It is clear, however, that the contents of the digestive ceca assume a deep
red color. The stem of the tail, with the exception of the caudal third
or fourth, is filled with large fluid globules that take the stain and become
a brick-red color. Young specimens do not take the stain at all, and at a
later stage the bodies of the larvae assume a diffuse pink or rose color.
The use of other stains, methylene blue, dahlia, pyronin, brilliant cresyl
blue, Janus green, light green, and methyl violet did not give significant
results on this cercaria. If very dilute solutions were employed the
staining was slight and diffuse, not differential, and when stronger solu-
tions were used the cercariae encysted very quickly.
The mouth opening is subterminal and the oral sucker measures from
0.06 to 0.08 mm. in diameter in living specimens. In fixed and stained
specimens the diameter is from 0.05 to 0.06 mm. There is a short
prepharynx, the pharynx measures from 0.02 to 0.027 mm. in diameter,
and the esophagus is long, extending about two-thirds of the distance to
the acetabulum, where it bifurcates to form the intestinal ceca (Fig.
4, In). The ceca end blindly about two-thirds of the distance from the
acetabulum to the caudal end of the body. The esophagus as well as the
ceca is lined with epithelium and this point is significant for life history
and taxonomic identification.
The excretory system (Fig. 2) has been worked out in detail and the
pattern confirmed on dozens of cercariae. The system forms as two
separate parts, one right and the other left, as described for other cer-
cariae. The longitudinal ducts fuse near the posterior end of the body
and this portion becomes the future excretory vesicle. With the con-
268 II. \V. STUXKARD AND C. RUTH SHAW
striction that forms the tail the posterior end of the vesicle is denoted
and the two excretory pores open on either side of the tail as shown in
Fig. 2. The details of the excretory system are shown in the figure.
The collecting ducts extend forward to the level between the pharynx
and oral sucker, passing on the ventral side of the intestinal ceca. The
anterior portions of these ducts contain excretory concretions. Each
collecting duct turns posteriad and this recurrent stem contains two large
ciliated areas. At the- level of the intestinal bifurcation the recurrent
ducts divide into anterior and posterior branches. Each of the anterior
and posterior branches bears three clusters of flame cells with three cells
in each cluster. The cone of cilia in a flame cell measures from 0.008
to 0.01 mm. in length.
The reproductive organs are represented by a mass of deeply staining
cells, the anlagcn of the gonads. situated in front of the excretory vesicle,
and a strand of cells which extends forward connecting with another
cell mass in trout of the acetabulum. The strand of cells will form the
^ono.lucts and the cell mass in front and sometimes to the left of the
acetabulum is the anlage of the genital pore and copulatory organs.
The cercari.T are produced in redire. A redia has two " feet," a
birth pore near tin- oral sucker, and an intestine which contains orange-
colored granules. In a small redia the germ- masses are situated in the
caudal third of the body and the posterior tip may be protruded in a tail-
like or foot-like protuberance that is used like the teet in locomotion.
There is a muscular lip-like snout in front of the oral sucker; the sucker
measure- from 0.05 to 0.07 mm. in diameter and in a young redia the
intestine extends through three fourths of the body length. Redia-
increase to a length of 2.1 mm. and a width of 0.4 mm. The small
rediie may have one or more fully formed ccrcaria- in their bodies and
large redia- contain from ten to thirty more or less developed ecrcari;e.
Ccrcaria sciisifcra belongs to the Megalura group of cercariae, out-
lined by Cort (1915) and extended by Si-well (1922). It agrees closely
with C. purpitrfc and C. patclhc, marine species described by Lebour
(1907, 1912) and indeed may be s|,,vilically identical with C. purpurcc.
Slight differences between the present specimens and the account of
Lebour in regard to the arrangement of the gland cells, the presence of
a " neck " region, and the relations of the excretory system and the tail.
make it impossible to determine with certainty whether the specimens
may be referred to C. pur put Vcordingly. a new name is proposed
for them with the understanding that it will disappear as a synonym if
further studies prove the American species to be identical with the
European. In the paper cited. Lebour ( 1912) predicted that the adult
stage of the parasite occurs in a bird. Two years later, Lebour (1914)
DILUTION OF SEA WATER AND CERCARI^ 269
compared young stages of Parorchis acanthus Nicoll with C. purpnrce
and identified the cercaria as the larval stage of that species, confirming
a prediction made by Nicoll some years previously. Believing that
Parorchis is closely related to the echinostomes, Lebour stated that the
second intermediate host is probably a mollusk. In a later report,
Lebour and Elmhirst (1922) reported that C. f>nrpur(C encysted in the
mantle of Cardiuni cdnlc and Mytilns cditlis. Their figure of the cyst
from side view indicates that it is on the surface rather than embedded
in the mantle of the host. The observations of Lebour and Elmhirst
appear to complete the life cycle of the species but the experiments were
not sufficiently controlled to exclude other possibilities. In the first and
only successful experiment reported, the cercaria? used " swam actively
by a strong side to side lashing of the tail." It is questionable whether
these larvae were actually C. purpurcc since megalurous cercarise do not
employ this method of swimming. Two tvpes of larval trematodes were
found encysted in the single specimen of Cardhtni used in the experi-
ment and the more abundant species was identified as Echinostomum
sccnndniii. Consequently, two types of cercarire were introduced unless
the intermediate host was already infected when the experiment was
begun. Apparently the authors did not know whether or not the bi-
valves used in the experiment were previously infected, and results of
such experiments are not at all conclusive. It C. purpura, like C. scn-
sifcra will encyst on any surface to which it adheres, it is only natural
that the larva? should encyst on the mantle of mollusks placed in aquaria
with them, and it does not necessarily follow that these mollusks are the
normal intermediate hosts.
The idea at once presents itself that C. scnsifcra is a larval stage of
Parorchis aritits Linton, 1914, an unusual trematode in which free
miracidia were found in the uterus and in which each miracidium con-
tained a well developed redia. Such a life cycle would explain the ob-
servation of Lebour that sporocysts or other preredial stages of C. pnr-
purcc were not found in the snail host. Linton (1928) gave a further
discussion of P. ai'itits and argued against the suggestion of Nicoll that
the American species is identical with Parorchis acanthus.
»
SUMMARY
The problem concerning the origin and distribution of closely related
parasites that occur in marine and fresh-water hosts is discussed. Mi-
gration of hosts and transfer to new hosts are the only explanations that
appear tenable. The essential factor is the ability of the free-swimming
larvre to live and remain infective in a changed environment. Experi-
ments on the ability of six species of marine cere-arise to withstand dilu-
270 II. \V. STUXKARD AND C RUTH SHAW
tion of sea water show that these larvae manifest normal activity for
considerable periods of time in solutions containing only one-eighth to
one- fourth sea water. The observations indicate that these cercarise
are able to complete their life cycles in brackish water and denote the
extent to which these organisms may migrate into brackish or fresh
water.
BIBLIOGRAPHY
Anoi.i'H, E. F., 1925. Some Physiological I Jistinctions Between Freshwater and
Marine Organisms. Biol. Bull.. 48: 327.
AI.LEE, W. C., 1923. Studies in Marine Ecology. III. Some Physical Factor-
Related to the Distribution of Littoral Invertebrates. Biol. Bull., 44: 205.
I'.iu.-H.Ar. K., AMI E. REISIX<;KR, 1928. Allgemeine Einleitung zur Natur-
geschichte cler Platyhelminthes. In Handhuch der Zoologie. Kiikenthal-
Krumbach. Berlin and Leipzig.
CORT. \\'. W.. 1915. Some North American Larval Trematodes. ///. Biol.
Monogr.. 1: 447.
CORT, W. W., 1918. Adaptability of Schistosome Larvrc to New 1 losts. Jour.
Parasit.. 4: 171.
COWLES, R. P., 1930. A Biological Study of the Offshore Waters of Chesapeake
Bay. Hull. Bur. Fish., 46: 277. '
FUJIKMAXX, ().. 1<;2<S. Trematoda. In Handhuch der Zoologie, Kukenthal-
Krumbach, Berlin and Leipzig.
HAY, O. P.. 1908. The Fosil Turtles of North America. Carnegie hist. IVash.,
I 'lib. No. 75.
LKBOUR, M., l'X)7. Larval Trematodes of the Northumberland Coast. Trans.
Nat. Hist. Soc. Newcastle, N.S.. 1: 437-^54, 500-501.
LEBOUR, M. V., 1912. A Review of the British Marine Cercariae. Parasit., 4: 416.
LEBOUR, M. V., 1914. Some Larval Trematodes from Millport. Parasit., 7: 1.
LEBOUR, M.. AND R. ELM HIRST, 1922. A Contribution Towards the Life-History
of Parorchis acanthus Nicoll, a Treinatode in the Herring Gull. Jour.
Mar. Biol. Ass. Plymouth, N.S., 12: 829.
LINTON, E., 1928. Notes on Treinatode Parasites of Birds. Proc. U. S. Nat.
Mus.. 73: 1.
Looss, A., 1902. Ueber neue und bekannte Trematoden aus Seeschildkroten.
Zool. Jalirh.. Syst.. 16: 411.
MACKIN, J. G., 1930. A New Pronocephalid Monostome from a Freshwater
Turtle. Jour. Parasit., 17: 25.
MANTER, H. W., 1926. Some North American Fish Trematodes. ///. Biol.
Monogr., 10: No. 2, p. 7.
MARSHALL, F. K., JR., AMI II. W. SMITH. 1930. The Glomerular Development
of the Vertebrate Kidney in Relation to Habitat. Biol. Bull., 59: 135.
MEIXNER, J., 1926. Bcitrag zur Morphologic und zum System der Turbcllaria-
Kbabclocoela. II. Zcitschr. t^iss. Bio!.. Abt. A., Morph. u. Okol.. 5: 577.
MILLER, H. M.. AND F. E. NORTH i T. 1(>26. The Seasonal Infestation .of Nassa
obsoleta (Say) with Larval Trematodes. Biol. Bull.. 50: 490.
NEEDHAM, J., 1930. On the Penetration of Marine Organisms into Freshwater.
Biol. Zentralbl. 50: 504.
NICOLL, W., 1915. A List of the Treinatode Parasites of British Marine Fishes.
Parasit., 7: 339.
NICOLL, W., 1924. A Reference List of the Treinatode Parasites of British Fresh-
water Fishes. Parasit.. 16: 127.
PANTIN, C. F. A.. 1931. The Adaptation of Gunda ulvse to Salinity, III. Jour.
/:.r/vr Biol., 8: 82.
DILUTION OF SEA WATER AND CERCARI^ 271
REISINGER, E., 1928. Allgemeine Einlcitung xur Naturgeschichte der Vermes
Amera. In Handbuch der Zoologie, Kiikenthal-Krumbach, Berlin and
Leipzig.
SEWEI.L, R. B. S., 1922. Cercarice indicaj. Ind. Jour. Mcd. Res., 10: Suppl. No.: 1.
STUNKARD, H. W., 1917. Studies on North American Polystomid;e, Aspido-
jiastricke, and Paramphistomidse. ///. liiol. Moiwtjr., 3: 285.
STUNKARD, H. W., 1929. The Parasitic Worms Collected by the American Mu-
seum of Natural History Expedition to the Belgian Congo, 1909-1914.
Hull. Am. Mus. Nat. Hist., 58: 233.
STUNKARD, H. W., 1930. Morphology and Relationships of the Trematode
Opisthoporus aspidonectes (MacCallum, 1917), Fukui, 1929. Trans.
.•liner. Micr. Soc., 49: 210.
STUXKARD, H. W., 1930</. The Life History of Cryptocotyle lingua (Creplin),
with Notes on the Physiology of the Metacercarise. Jour. Morph. and
PhysioL, 50: 143.
STUXKARD, H. W., I930b. An Analysis of the Methods used in the Study of
Larval Trematodes. Parasit.. 22: 268.
STUXKARD, H. W., 1930t\ The Effect of Dilution of Sea-Water on the Activity
and Longevity of the Cercariae of Cryptocotyle lingua. Anat. Rcc., 47:
362.
STUXKARD, H. 4V., AXD C. H. ALVEY, 1930. The Morphology of Zalophotrema
hepaticum, with a Review of the Trematode Family Fasciolidrc. Parasit.,
22: 326.
STUXKARD, H. W., AXD R. F. NIGRELLI, 1930. On Distomum vibex Linton, with
Special Reference to its Systematic Position. Biol. Bull., 58: 336.
TEXXEXT, D. H., 1906. A Study of the Life History of Bucephalus haimeamis ;
a Parasite of the Oyster. Quart. Jour. Micr. Scl., 49: 635.
WILLISTOX, S. W., 1914. Water Reptiles of the Past and Present. Univ. of
Chicago Press, Chicago, 111.
\\"OODHEAD, A. E., 1929. Life History Studies on the Trematode Family
Bucephalidae. Trans. Ainer. Micr. Soc., 48: 256.
WOODHEAD, A. E., 1930. Life History Studies on the Trematode Family
Bucephalidse. No. II. Trans. Anier. Micr. Soc., 49: 1.
ZITTEL, K. A.. AXD C. R. EASTMAN, 1913. Textbook of Paleontology, 2d ed.,
Vol. 1.
Vol. LXI, No. 3 December, 1931
THE
BIOLOGICAL BULLETIN
•
PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY
THE TENSION AT THE SURFACE OF MARINE EGGS,
ESPECIALLY THOSE OF THE SEA URCHIN,
ARBACIA
E. NEWTON HARVEY
(From the Marine Biological Laboratory, Woods Hole, Mass., and the
Physiological Laboratory, Princeton University')
Biologists frequently speak of the surface tension of cells, com-
paring their form, movements and division to phenomena connected
with surface tension at oil-water interfaces. It seems unlikely that
the tension is a true surface tension between non-misceable fluids, but
the behavior of deformed spherical cells shows very clearly that a
surface force exists which can best be referred to as the " tension at
the surface " without implying either elastic tension or surface tension.
Estimates of its magnitude have been made in different ways and until
recently the values obtained have been relatively high, from 10 to 50
dynes per centimeter. I believe the tension is very much smaller than
this in many cells.
In a recent paper (1931) I have described an approximation method
for determining the tension at the surface of an unfertilized egg in sea
water, from the centrifugal force necessary to pull the egg apart. In
the worm, Chatopterus pergamentaceus, the forces necessary are small
and the whole process can be observed and photographed in the micro-
scope-centrifuge (Harvey and Loomis, 1930), whose maximum speed is
4000 R.P.M. The value obtained for Chatoptcrus was about one dyne
per centimeter, which represents the maximum value, since all the as-
sumptions made were such as to give a maximum. The true value is
probably considerably less than this, but the fact that the surface forces
are so low is a point of great interest.
The method does not allow us to decide whether this force is a true
surface tension at a liquid-liquid interface or the elastic tension of a
membrane, because we cannot tell whether or not the strain is inde-
pendent of the stress. Micro-dissection studies (Chambers, 1921) in-
dicate that the surface of marine eggs is surrounded by an actual con-
sistent film variously spoken of as the " pellicle " or vitelline membrane,
19 273
274 E. X. HARVEY
which lifts i iff and hardens to form the fertilization membrane of Ar-
bacia eggs. Since a ////'// elastic film will take the same configuration
under distortion as a surface showing true surface tension, and since
the pellicle in the egg becomes discontinuous at the time of cell division
or when the egg is fragmented by ccntrifuging, we may regard such a
film as having the properties of a true surface (except for the relation
In -i ween stress and strain), and on this basis calculate its tension by
methods which at least give order of magnitude and a maximum value.
The argument for the Arbacia egg is somewhat different from that
used in the case of Chcetopterus, because .-Irbacia pulls apart in a quite
different manner. In Chcetopterus in sea water an oil spherule pulls
away from the rest of the egg at 4000 R.P.M. (11 cm. radius), re-
maining attached to the egg by a long stalk pulled out to many times
tin- diameter. The picture is the same if the eggs are suspended in a
sugar solution of the same density or of greater density, when the eggs
float. In the latter case a yolk spherule is pulled away from the re-
mainder of the egg, again with a long connecting stalk.
Unfertilized eggs of the sea urchin, Arbacia punctulata, cannot be
rapidly pulled apart with this force but, at about 7000 R.P.M. (same
radius), if suspended in a medium of the same density as the egg, they
< longate, form dumb-bell shapes and in 4 minutes separate into a lighter
and a heavier half of nearly the same size. The egg may be regarded
as a sphere pulled into a cylinder with rounded ends by the buoyant
force of tin- oil and the weight of the heavier yolk mass. In this process
tlit surface area increases about 25 per cent.
It is well known that a cylinder of fluid becomes an unstable form
when its length exceeds its circumference (2irr), i.e., when its length is
about three times its diameter (Lord Rayleigh, 1879). Under these
circumstances it will divide into two. The surface tension (a) around
the circumference should then just balance the forces pulling the
cylinder apart, altrr = forces stretching cylinder.
I f we regard the egg as a sphere non-misceable with sea water, we
ran calculate its circumference when drawn into a cylinder with hem-
ispherical ends whose length is -n- times its diameter. The breaking up
into two spheres of such a form will only be delayed because of the
viscosity of the sphere.
We wish to know the radius of a cylinder of height, //, with hem-
i^pheres at each end of radius, r, in which h-\-2r = 2irr and whose
volume equals that of a sphere, the . Irhacia egg, of diameter, d.
Hen. . . irr-h + 4/3*r* = 4/3*(d/2) 3
But, h = v2r - - 2r
Substituting, irr-(ir2r--2r) + 4/3^= --4/3ir(d/2)3
or 27rr3 — 27r;-3 -- 4/37r;-3
SURFACE TENSION OF MARINE EGGS
Since the average diameter of an Arbacia egg is 74 p.,
275
r =23.0
,*.
The forces pulling the egg apart arc due to the weight of the heavy
fragment (//) and the buoyancy of the light fragment (L), which can
be determined from the volume of the fragments (V), their densities
(/o), and the density of the medium (p M} which is equal to the density
of the whole egg since the eggs are centrifuged in a medium of equal
density.
Force (in dynes) = = V,, (PII—PM) + VL (PM — pL} X 980 )( C,
where C- = centrifugal force in terms of gravity.
Since the egg pulls into approximately equal parts, the density of the
light fragment must be as much less than the medium as the density of
the heavy fragment is greater than the medium.1 If V E is the volume
of the egg, the whole relation therefore becomes :
2m=rE (pM—PL)9SOXC. (1)
It is only necessary to know the diameter of an egg, the density of
the medium and that of the light fragment and the centrifugal force to
divide the eggs.
The diameter of the Arbacia egg (74ju) gives a volume (V E} of
2.12 X 10~7 cm.3 To get the density of the eggs (without jelly) they
must be suspended in a medium of the same osmotic pressure as well
as the same density. Lucke has found that cane sugar of .95 moial
concentration (342 grams cane sugar added to 1 liter water) causes
neither swelling nor shrinking of Arbacia eggs.2 I find that when cen-
trifuged in one part of sea water and 3 parts of .95 M cane sugar, some
1 Dr. Balduin Lucke, in the course of some experiments on the osmotic prop-
erties of yolk and clear halves, has measured the volumes in cubic micra and com-
pared the sum of these volumes with the volume of the original egg. Each figure
is the mean of 50 cells and must be multiplied by 100.
Date
Control egg
A
Colorless half
B
Yolk half
C
Sum of
B and C
Aug. 7
1895
1013
939
1952
8
2088
'1185
932
2117
" 13
1954
1129
814
1943
" 14
1905
983
904
1887
" 15
2030
1120
945
2065
The yolk half is somewhat smaller than the colorless half, averaging around
11 per cent smaller.
- Private communication. See also Lucke, 1931. Biol. Bull., 60: 75.
276 K. X. HARVEY
lots of eggs float, most sink and some remain suspended even under
high centrifugal forces. This mixture has a density of 1.081 at 23° C.s
In one part sea water and four parts .95 M cane sugar the eggs of most
female-; float; others sink very slowly. Its density is 1.085 at 23° C.
\Yc may consider the density of the medium (p.v) and the egg to be
1.083. I believe Heilbrunn's (1926) value of 1.0485 (12.5 per cent
-ugar) and 1.0656 (16.5 per cent sugar) for different lots of eggs are
too low, because he suspended the eggs in pure sugar solutions which
were hypotonic and volume changes must have occurred. However,
the density of different lots of eggs docs vary considerably. Eggs with
jelly have a density of about 1.090, whereas without jelly the density is
about 1.083 to 1.084.
To obtain the density of the light fragments (PL), the eggs {without
jelly] are centrifuged in the sugar-sea water mixture of the same density
and the light fragments transferred to other mixtures of sugar and sea
water. They mostly float in 10 parts of sea water to 20 parts .95 M
>ugar, whose density is 1.076, whereas they all sink in 12 parts sea
water to 20 parts .95 M sugar, whose density is 1.073. We may there-
fore consider their density to be 1.075 and p u -- PL '- - -008.
The sugar-sea water mixture of the same density as the egg is not
toxic. Although eggs will not fertilize in the sugar solution, they can
be fertilized and develop normally when removed to sea water after an
immersion of five hours.
In determining the centrifugal force the time factor is an important
consideration. Part of the time is involved in the separation of gran-
ules of different density within the eggs. Only when this happens do
the stretching forces appear. I 'art of the time is connected with the
slow pinching of tin- egg in two. Kvcn when an egg has assumed a
prolate spheroid shape, recovery of the spherical form is very slow, a
matter of many minutes. Dumb-bell-shaped eggs do not pinch in two
after the centrifugal force is removed but remain dumb-bells for many
minutes, gradually becoming spherical again (after 40 minutes). There
is also considerable variation in the ease with which eggs can be pulled
apart. Eggs from some females fragment at 122 r.p.s. in 4 minutes,
others do not but will fragment in 12 minutes. Eggs which fragment
in 4 minutes at 122 r.p.s. are not pulled apart in 4 minutes at 112 r.p.s.,
but fragment in 12 minutes. The same eggs centrifuged for 20 minutes
at 100 r.p.s. do not fragment but do in 24 minutes. The eggs of some
females pulled apart at 60 r.p.s. in 30 minutes, but not at 50 r.p.s. in 90
minutes.
3 Densities were determined with a hydrometer calibrated for 15° C./15" C.,
reading to the third decimal place. The temperature correction will be small.
KDITOR'S NOTE:
The equation on page 277 of the article by E. N. Harvey in the
December, 1931, issue should be corrected to read as follows:
2.12 x iQ-7 x .8 ) ; IQ-- ) : io3 ) : 1.6 > : io3
6.28 X 23 X IO-4
^•i^ A iu • x -a x iu~- ; iu- x i.o x iu°
* = 6.28 X 23 X 10- =0,19 dynes per cm.
The result is very much lower than the 10 to 25 dynes per centimeter
observed by Vies (1926) for the egg of another sea urchin, probably
Paracentrotus.
Should the eggs rest on a surface so that only the buoyant force of
the light half is operative, the value will be about one-half of the above.
The stretching forces must act against not only a tension at the surface,
but viscous forces of the egg as well, which again will lower the figure.
Finally, the calculation is based on the view that the tension is a true
surface tension. If the elastic tension of a pellicle is involved, we are
observing its breaking strength and its tension must be considerably
less for a given stretch. It must be emphasized that 0.2 dyne per
centimeter is not a very accurate value, but a maximum one, and again
illustrates the very low tension at the surface of eggs presumably sur-
rounded with a pellicle. The question arises as how general this order
of magnitude is for other marine eggs.
Dr. H. K. Hartline has pointed out to me that fluid spheres from
which a small oil spherule is drawn out should become unstable when
the neck connecting oil spherule with the sphere has the same diameter
as the spherule. The spherule will then pinch off as a separate drop.
In this case the surface tension around the circumference of the oil
spherule should counterbalance the buoyant force of the oil. In the
case of Chcctoptcrus eggs, the separation of the oil spherule occurs only
after a long stalk has been pulled out and the buoyant force of the oil
was regarded as counterbalancing the tension around the circumference
of the stalk, 9/x, in diameter (Harvey, 1931). If the circumference of
the oil spherule is considered, 34 /x, in diameter, the value for the tension
at the egg surface comes out about one-quarter of 1.32 dynes, or .33
dynes per centimeter.
about 1.083 to 1.084.
To obtain the density of the light fragments (p/,), the eggs (without
jelly) are centrifuged in the sugar-sea water mixture of the same density
and the light fragments transferred to other mixtures of sugar and sea
water. They mostly float in 10 parts of sea water to 20 parts .95 M
sugar, whose density is 1.076, whereas they all sink in 12 parts sea
water to 20 parts .95 M sugar, whose density is 1.073. We may there-
fore consider their density to be 1.075 and pM --pL - = .008.
The sugar-sea water mixture of the same density as the egg is not
toxic. Although eggs will not fertilize in the sugar solution, they can
be fertilized and develop normally when removed to sea water after an
immersion of five hours.
In determining the centrifugal force the time factor is an important
consideration. Part of the time is involved in the separation of gran-
ules of different density within the eg^. ( Mily when this happens do
the stretching forces appear. Part of the time is connected with the
slow pinching of the egg in two. Even when an egg has assumed a
prolate spheroid shape, recovery of the spherical form is very slow, a
matter of many minutes. Dumb-bell-shaped eggs do not pinch in two
after the centrifugal force is removed but remain dumb-bells for many
minutes, gradually becoming spherical again (after 40 minutes). There
is also considerable variation in the ease with which eggs can be pulled
apart. Eggs from some females fragment at 122 r.p.s. in 4 minutes,
others do not but will fragment in 12 minutes. Eggs which fragment
in 4 minutes at 122 r.p.s. are not pulled apart in 4 minutes at 112 r.p.s.,
but fragment in 12 minutes. The same eggs centrifuged for 20 minutes
at 100 r.p.s. do not fragment but do in 24 minutes. The eggs of some
females pulled apart at 60 r.p.s. in 30 minutes, but not at 50 r.p.s. in 90
minutes.
3 Densities were determined with a hydrometer calibrated for 15° C./15" C.,
reading to the third decimal place. The temperature correction will he small.
SURFACE TENSION OF MARINE EGGS 277
If eggs which have been pulled into dumb-bell form at 120 r.p.s. are
observed in the microscope centrifuge at 60 r.p.s., a few pull apart into
two spheres. The connecting strand between the spheres does not be-
come long and fine as in the pulling off of an oil spherule in Chcetop-
tcrus, but the break can be observed to occur when the length is about
three times the diameter of the resultant half egg.
Since the centrifugal force (C) in terms of gravity is given by
C = .443;r, for 11 cm. radius, where n = = revolutions per second, C is
6380 for 120, 5560 for 112, 4430 for 100, and 1595 for 60 r.p.s.
Selecting a speed of 60 r.p.s., and inserting in (1), we have:
2.12 X 10"7 X .8 X 10-2 X 103 X 1.6 X 103
ffs= 6.28 X 23 X 10- ^.19 dynes per cm.
The result is very much lower than the 10 to 25 dynes per centimeter
observed by Vies (1926) for the egg of another sea urchin, probably
Paracentrotiis.
Should the eggs rest on a surface so that only the buoyant force of
the light half is operative, the value will be about one-half of the above.
The stretching forces must act against not only a tension at the surface,
but viscous forces of the egg as well, which again will lower the figure.
Finally, the calculation is based on the view that the tension is a true
surface tension. If the elastic tension of a pellicle is involved, we are
observing its breaking strength and its tension must be considerably
less for a given stretch. It must be emphasized that 0.2 dyne per
centimeter is not a very accurate value, but a maximum one, and again
illustrates the very low tension at the surface of eggs presumably sur-
rounded with a pellicle. The question arises as how general this order
of magnitude is for other marine eggs.
Dr. H. K. Hartline has pointed out to me that fluid spheres from
which a small oil spherule is drawn out should become unstable when
the neck connecting oil spherule with the sphere has the same diameter
as the spherule. The spherule will then pinch off as a separate drop.
In this case the surface tension around the circumference of the oil
spherule should counterbalance the buoyant force of the oil. In the
case of Chfftopterus eggs, the separation of the oil spherule occurs only
after a long stalk has been pulled out and the buoyant force of the oil
was regarded as counterbalancing the tension around the circumference
of the stalk, 9 p. in diameter (Harvey, 1931). If the circumference of
the oil spherule is considered, 34 //, in diameter, the value for the tension
at the egg surface comes out about one-quarter of 1.32 dynes, or .33
dynes per centimeter.
Jto
E. N. HARVEY
Similar reasoning applied to the fertilized egg of the mollusk, Ill\-
anassa obsolete, in which the oil can be observed to pull off as with
Cheetopterus in the microscope centrifuge, gives a value of 1.1 dynes
per centimeter.
The eggs of the mollusk, Cumingia tclllnoidcs, in sea water behave
differently. They pull out into long cylinders 5 to 8 times their width,
when the oil separates as a spherule which floats to the surface. Other
clear spherules may separate also. The oil spherules are about 25 p. in
diameter while the original egg is some 62 p. in diameter. If we as-
sume that unstable conditions appear when the egg is pulled out to a
cylinder whose diameter equals that of the oil spherule, we can calculate
roughly the tension at the surface as follows:
cnrd= VO(PW~ ~ PO) X y X C,
where a = = tension at surface, d- = diameter of oil spherule, p,,--- den-
sity sea water, 1.025, p0 == density of oil, C = centrifugal force in terms
of gravity, y, and V0 = volume of the oil. Assuming the density of
the oil to be .925 and observing that these eggs pull apart in 8 minutes
at 122 r.p.s. (11 cm. radius), we have
78.5 X 10-4cr=---6.42 X 10-° (1.025 — .925) X 103 X 6.6 X 103
a- .54 dynes per centimeter.
Again the tension at the surface comes out a low value.
The unfertilized egg of the worm, Xcrcis, possesses a definite mem-
brane. Its granules cannot be stratified by easily attainable centrifugal
forces. However, the fertilized egg of Xcrcis can be stratified but can-
not be fragmented, even at 17000 times gravity, although a slight
tendency to elongate occurs. The membrane of Arcrcis is very strong.
Calculations from the amount of oil (assuming its density = .925)
present in the Xcrcis egg indicate that the tension to withstand 17000
times gravity must be considerably greater than 24 times that of the
egg of Chcetoptcrus.
SUMMARY
Calculations from the centrifugal force necessary to pull an Arbacia
egg into two nearly equal parts, a yolk half and a clear half, indicate
that the tension at the surface for 25 per cent increase in area is less
than 0.2 dyne per cm., with considerable variation in different eggs.
Similar calculations based on the force necessary to pull an oil
spherule away from the remainder of the egg give maximum values of
0.33 dynes per centimeter for Cliactoptcrns. 1.1 dvnes per centimeter
for Illyuassa and 0.54 dyne per centimeter for Cumingia.
SURFACE TENSION OF MARINE EGGS 279
LITERATURE CITED
CHAMBERS, R., 1921. Biol. Bull., 41: 318.
HARVEY, E. N., 1931. Biol. Bull., 60: 67.
HARVEY, E. N., AND A. L. LOOMIS, 1930. Science, 72: 42.
HEILBRUNN, L. V., 1926. Jour. Exper. Zoo!., 44: 255.
LORD RAYLEIGH, 1879. Proc. Roy. Soc., 29: 83.
VLES, F., 1926. Arch. d. Physique Bio!., 4: 263.
THE THEORY OF MITOGENETIC RADIATION
G. WELLFORD TAYLOR AND E. XEWTOX HARVEY
(From the Physiological Laboratory, Princeton University)
INTRODUCTION
It has long been supposed that nuclear and cell division may be af-
fected by external as well as internal factors. While it is true that
increase in temperature will accelerate division rate and many means
are known to retard or prevent cell division, the evidence that mitosis,
except in the case of special tissues, can be initiated or accelerated by
definite compounds or by any means (except heat) is far from con-
vincing. This is particularly true in the cast.' of mitogenetic rays of
Gurwitsch (1923), rays given off by cells or cell extracts that will in-
duce division in another cell.
This theory, which lias many opponents as well as advocates, is set
forth at some- length in (iurwitsch's monograph (1926) and an article
in Protoplasma (1929), recently reviewed by Hollaender and Schoeffel
(1931). It is based upon the following observations: if an onion root
is placed vertically so that the meristematic area is perpendicular to a
second, horixontally placed, root, at a distance of not more than 4 cm.,
and left in this position for at least 20 minutes, then, when sections are
made of the first root, it can be seen on counting the number of dividing
cells in the two halves of this root that cell division was markedly stimu-
lated in the area exposed to the second root. The induced increase in
cell division in such an experiment may be as great as 80 per cent. The
area affected is always small, extending rarely for more than 50 micra.
This mitogenetic effect is equally as pronounced whether the roots are
in air or in water during the induction; the effect is not markedly
lessened by the interposition of a sheet of quart/ between the " sender "
and "detector" roots, but is completely obliterated it a sheet of glass
is placed between the two roots. The further observation that this
mitogenetic influence, besides being propagated in a rectilinear manner,
was capable of regular rectilinear reflection led (iurvvitsch and Frank
(1927) to conclude1 that this influence, or M-ray, was in nature identical
with ultra-violet light. By comparing the induction effect of one root
on another with the mitogenetic effects of ultra-violet light of varying
wave lengths on a similar root, a wave length of 1 ''00-2300 angstroms
has bi-cii ascribed t<> the M-ray. This conclusion was influenced to some
280
MITOGENETIC RADIATION 281
extent by the observations of Rawin (1924), who has shown that, while
the M-rays are effective in inducing cell division for a distance of very
nearly 4 cm., they can only penetrate glass for a distance of 50 micra
before being completely absorbed.
Reiter and Gabor (1928), in a long series of similar studies on the
reflection, refraction, diffraction, and absorption of the mitogenetic
rays, find that they show the same properties in respect to these phe-
nomena as ultra-violet light with a wave length of 3300 to 3400 ang-
stroms, with a second maximum at 2800 A°.
Frank (1929) ascribes this difference in the assigned wave length
of the M-rays to the two sets of workers having used different inten-
sities of ultra-violet light in making their comparative studies. In spite
of their difference of opinion over the wave length of the M-rays, they
all agree that they are identical with ultra-violet light.
The experiments of Gurwitsch received almost immediate confirma-
tion by a group of Russian workers and somewhat later by a smaller
group of non-Russian workers. These have extended the known
sources of mitogenetic rays until it includes the following tissues : root
tips and other embryonic plant structures, Gurwitsch (1923), Gurwitsch,
A., and N. (1924), Rawin (1924), Frank and Salkind (1926), Wagner
(1927), Baron (1926), Reiter and Gabor (1928), Borodin (1930),
Stempell (1929), Hollaender and Schoeffel (1931) ; potato leptom, Kis-
liak Stratkewitsch (1927) ; twenty-four-hour-old sterile beet pulp, Anna
Gurwitsch, as quoted by Gurwitsch (1929) ; tadpole heads, Gurwitsch,
A., and L. (1925), Rusinoff (1925), Reiter and Gabor (1928); brain
of young tadpole, Anikin (1926) ; bacteria, J. and M. Magrou (1927,
1928), J. and M. Magrou and Choucroun (1929), Baron (1926, 1928),
Borodin (1930) ; yeast,. Baron (1926), Gurwitsch (1926, 1929), Reiter
and Gabor (1928), Borodin (1930), Hollaender and Schoeffel (1931) ;
sea urchin eggs, Frank. and Salkind (1927), Salkind (1929), Frank and
Kurepina (1930); animal half of amphibian morula, Anikin (1926);
yolk of chicken egg, Sorin (1928) ; corned epithelium of starved rats,
triton, and frog, Gurwitsch, L., and Anikin (1928) ; contracting muscle,
Siebert (1928), Frank and Popoff (1929) ; isolated frog heart, Salkind,
Potozky, and Zoglina (1930) ; Jenson sarcoma, Siebert (1928) ; malig-
nant tumors, Reiter and Gabor (1928), Hollaender and Schoeffel
(1931) ; bone marroiv, spleen and lymph glands of young rats, Suss-
monowitsch, quoted from Gurwitsch (1929) ; confirmed for bone mar-
roz\j by Siebert (1928) ; active isolated nerves, Wassiliew, Frank, and
Goldenberg (1931); reabsorption processes accompanying amphibian
metamorphosis, Blacher (1930), Blacher and Bromley (1930), Blacher
and Holzmann (1930), Bromley (1930) and Holzmann (1930); both
G. W. TAYLOR AND E. N. HARVEY
normal and hemolysed blood of frog and rat, Gurwitsch, A. and L.
(1926). and Sorin (1926) ; blood and urine of healthy persons, Siebert,
W. \Y. ( 1930), Gurwitsch, A. and L. (1928), Hollaender and Schoeffel
(1931), Gesenius, H. (1930), Potozky and Zoglina (1928).
The M-rays arc thought to take their origin in some oxidative re-
action connected with normal metabolic processes of the tissue emitting
the ray and not necessarily to cell division. Frank and Popoff (1929)
attribute the rays emanating from muscular contraction to the explosive
decomposition of glycogen to lactic acid, while Siebert (1928) thinks
they originate in the oxidation of lactic acid. The latter to prove his
point has constructed several chemical models, oxidative reactions in
test-tubes, which imitate the radiating properties of living tissues.
Gurwitsch (1924, 1925), however, attributes the origin of the rays
to an enzymatic reaction similar to the oxidation of luciferin by lu-
cif erase. In the latter reaction visible light is emitted, while in the
former there is an emission of ultra-violet-like M-rays. To prove his
]M'int, Gurwitsch has succeeded in extracting from a pulp of onion roots,
by following the procedure of Dubois, two substances which are sep-
arately inactive, but which on being mixed will emit M-rays for as long
as an hour. He calls one of these substances " mitotin," and the other,
which he thinks an enzyme, " mitotase."
So far the proof of the existence of a mitogenetic radiation has been
entirelx physiological and rests upon the ability of the rays to induce an
increased cell division in the meristem of onion roots, in yeast and bac-
terial culture^. e.^s and other tissues with by far the greater emphasis
being placed on the onion root as a detector of the radiation.
The Gurwitsch school claims that normally there is a radial sym-
metry in the distribution of mitoses in an onion root, and that there is
never in a normal root a deviation of more than 10 per cent from this
symmetry.
The chief proof of the existence of the mitogenetic radiation is its
ability to destroy this symmetry by increasing by 20 to 80 per cent the
number of dividing cells in the area exposed to tin- radiation.
A ^ood many of the workers on these rays have been content to ac-
cept the assertion of the Gurwitsch school as to the existence of this
symmetry (c.<j., Borodin, 1930) and rather than undertake the laborious
and. extremely tedious task of counting the dividing cells in normal
roots in addition to making the necessary omuls for their experimental
roots, have, been publishing the results of experiments that were very
inadequately controlled. With these workers any variation from the
" normal " svimiietrv that was greater than 10 per cent was all too apt
to be regarded as being c\idcnce of a mitogenetic inllnence.
MITOGENETIC RADIATION
Obviously as soon as any doubts arc cast upon the actuality or uni-
versality of this radial symmetry the mass of evidence based entirely on
the existence of such a symmetry becomes questionable. Such is the
case at present, for in the recent work of Schwarz (1928), Rossmann
(1928), and von Guttenberg (1928, a and b), the existence of such a
symmetry as claimed by the Gurwitsch school is emphatically denied.
Schwarz in eight experiments in which he used onion roots as
sources as well as detectors of the M-rays obtained positive results of
15.6 per cent, 12 per cent, and 0.8 per cent respectively in 3, and nega-
tive effects of 2.2 per cent, 5.4 per cent, 5.5 per cent, 17.7 per cent, and
22.7 per cent in the other five experiments.
Rossmann and von Guttenberg in studying normal unexposed roots
found a variation from symmetry as high as 32 per cent in onion roots,
and 38 per cent in pea roots. In a large number of experiments they
were unable to obtain a mitogenetic effect greater than this variation
which they found in their normal untreated roots.
Gurwitsch (1928, a and &), in his reply to the criticism of the above
workers, reaffirms his conclusion that normally there is no more than
10 per cent variation. He criticises the technique of Rossmann in his
focussing of the sender root on the detector root, and in his handling
of the roots after treatment.
It is of significance that Wagner (1927) was only able to obtain a
mitogenetic effect from exposing one onion root to another when the
detector root contained relatively few dividing cells. With these small
numbers even a slight difference when expressed as a percentage dif-
ference would seem large.
Additional negative evidence has been offered by Choucroun (1930),
Urbanowicz (1927), and Rossmann (1928-29). Choucroun was un-
able to duplicate in later experiments the results obtained by J. and M.
Magrou and himself in the work referred to above (1929). He con-
cludes that the abnormalities appearing in sea urchin eggs on exposure
to a bacterial culture, and formerly attributed to a mitogenetic radia-
tion from the bacteria, were due to an actual passage of some substance
from the culture to the dishes containing the eggs. No effect could be
obtained when the vessels containing the eggs were tightly sealed.
Urbanowicz could not increase the rate of division in Paranicciuin by
exposing them to onion roots. Rossmann, in answer to Gurwitsch's
criticism of his earlier work, performed 47 experiments in which yeast
and onion roots were used both as senders and detectors of the M-rays.
In only one of the 47 experiments did he obtain a positive effect.
The lack of agreement between the Gurwitsch school on the one
hand and Schwarz, Rossmann, and von Guttenberg on the other as to
ro^V
LU| LIBRARY
284 G. W. TAYLOR AND E. N. HARVEY
the symmetry of the onion root, and the rather inconclusive experiments
of Wagner led us to study the distribution of dividing cells in the
meristem of several onion roots which had not been exposed to any
supposed source of a mitogenetic radiation.
EXPERIMENTAL,
The roots, grown in the laboratory to a length of 2-10 millimeters,
were fixed in Bouin's as modified by McClung. Cross-sections were
made 7-12 micra in thickness. They were stained with iron hema-
toxylin. Only those series that had been cut symmetrically were used.
This was to insure that any asymmetry observed could not be attributed
to an asymmetrically cut section.
In making our counts the number of cells in each quadrant of the
section was counted separately, then by adding the numbers obtained in
two adjacent quadrants one-half of each section could be compared with
the other, thus making it possible to compare the distribution of mitoses
on the two sides of each of two diameters for each section of the root.
A cell was considered to be in mitosis from the earliest recognizable
spireme stage until the two daughter cells had been completely sep-
arated. In the following tables the halves 1+2 are compared with
3 + 4, and 1 -f- 4 with 2 + 3, the location of the quadrants on the
section being indicated in the small circle at the top of the table by
the numbers 1, 2, 3, and 4.
Table I shows the distribution of dividing cells as noted in one root,
counted for 33 sections in the center of the meristematic area. When
the halves 1+2 and 3 + 4 are compared there is a maximal variation
of 31 per cent in one direction and 27 per cent in the other. Nineteen
out of the 33 sections counted showed a variation of more than 10 per
cent. When the halves 1+4 and 2-1-3 are compared 14 of the 33
sections show an asymmetry of more than 10 per cent. The maximal
variation is 14 per cent in one direction and 31 per cent in the other.
A second root counted for only 12 sections, in the center of the
dividing area, was just as asymmetrical as the first. Thinking that this
a-\ mme.try might have been due to an asymmetrical exposure of the
growing rool to li.^ht, a root that had been grown in complete darkness
was counted.
"Table II shows the distribution of mitose^ as noted in this root that
had been i^ro \\-n in absolute darkness. It was just as asymmetrical for
the 21 sections counted as either of the other two. When the halves
1+2 and 3 - - 4 are compared 9 sections vary from symmetry by more
than 10 per cent, the maximal variation being 24 per cent in one direc-
tion and JN per i nit in the other. When the halves 1+4 and 2 + 3
MITOGENETIC RADIATION
285
are compared 12 sections are more than 10 per cent asymmetrical, the
maximal variation being 5 per cent in one direction and 54 per cent in
the other. If sections containing only a few dividing cells, as at either
end of the meristematic area, are included the percentage differences may
be even higher.
On the basis of the observed asymmetry in normal onion roots by
Rossmann, von Guttenberg, and ourselves we cannot help but agree with
the first-mentioned workers and with Schwarz, that the existence of a
TABLE I
Series 38, sections cut 12 micra.
Sect. No.
No.
Mitoses
half 1, 2
Mitoses
half 3. 4
Diff.
Diff.
Mitoses
half 1 , 4
Mitoses
half 2, 3
Diff.
Diff.
per cent
per cent
18
107
94
13
12
95
106
- 11
- 10
19
143
130
13
9
152
121
31
20
20
100
118
- 18
- 15
117
101
16
10
21
96
137
- 41
- 30
109
124
- 15
- 12
22
125
106
19
15
124
107
17
13
23
90
80
10
11
90
88
2
2
24
119
173
- 54
- 31
119
130
- 11
- 9
25
72
91
- 19
- 21
77
87
- 10
- 12
26
105
134
- 29
- 22
115
93
22
19
27
94
120
- 26
- 22
94
109
- 15
- 14
28
106
99
7
6
116
97
19
16
29
126
157
- 31
- 20
126
123
3
2
30
134
98
30
27
134
112
22
16
31
86
86
0
0
86
87
- 1
- 1
32
113
91
22
19
113
80
23
29
33
82
75
7
8
82
71
11
13
34
97
95
2
2
97
102
- 5
- 5
35
116
114
2
2
116
108
8
7
36
103
89
14
13
103
79
24
23
37
100
124
- 24
- 19
100
114
- 14
- 14
38
134
128
6
4
134
130
4
3
39
98
130
- 32
- 25
117
111
6
5
40
133
121
12
9
126
128
- 2
- 2
41
109
126
- 17
- 13
123
112
11
9
42
132
116
16
12
119
129
- 10
- 8
43
160
136
24
15
152
144
8
5
44
120
113
7
6
138
95
43
31
45
98
113
- 15
- 13
104
108
4
- 4
46
135
126
9
7
133
128
5
4
47
121
127
- 6
~~ J
117
131
- 14
- 10
48
119
133
- 14
- 10
128
124
4
3
49
122
137
- 15
- 11
129
130
1
1
50
145
158
- 13
- 8
162
141
21
13
Totals:
3760
3876
- 116
- 3
3991
3645
346
9
286
G. W. TAYLOR AND E. N. HARVEY
mitogenetic radiation cannot be considered proved from the work done
on onion roots.
Since it has been agreed that the M-rays are identical with ultra-
violet light of a wave length of either 1900-2300 A* (Gurwitsch) or
3400 A° and 2800 A° (Reiter and Gabor), the rays should be able to
affect a photograph plate, regardless of their intensity, if a long enough
e\po>ure was given, fur the photographic plate is able to summate suc-
TABLE II
Series 152, sections cut 7 micra.
No.
No.
Mitoses
1, 2
Mitoses
half 3, 4
Diff.
Diff.
Mitoses
half 1, 4
Mitoses
half 2,3
Diff.
Diff.
per cent
per cent
59
74
70
4
5
70
74
- 4
- 5
60
22
29
— 7
- 24
35
16
19
54
61
32
23
9
28
33
22
11
33
62
41
31
10
24
41
31
10
24
63
35
37
- 2
5
38
34
4
10
64
52
49
3
6
57
44
13
23
65
40
36
4
10
42
34
8
19
66
43
51
- 8
16
46
48
- 2
4
67
56
54
2
4
55
55
0
0
68
53
44
9
17
51
46
5
10
69
72
62
10
14
66
68
_ ?
- 3
70
76
61
15
19
67
70
- 3
4
71
78
56
22
28
67
67
0
0
72
85
80
5
6
92
73
19
21
73
81
73
8
9
86
68
18
21
74
66
61
5
7
70
58
12
17
75
75
71
4
5
81
65
16
20
76
66
61
5
7
73
54
19
26
77
80
86
6
7
95
72
13
14
78
55
57
2
4
68
44
24
35
79
94
114
- JO
18
106
102
4
4
1 <>tuls:
1276
1206
70
5
1339
1145
194
14
<e-,sive small amounts of light striking it. But although practically
every worker in this field has tried to get such a photographic effect, all
have failed in their attempts with the exception of Reiter and Gabor
(1928) and these two workers suggest quite candidly that their results
should be confirmed.
In an effort to confirm their results, and to establish on a purely
phy>ical basis the existence of a mitogenetic radiation, and if possible to
determine more precisely the wave length of the M-ray, we have carried
out three series of experiments in an attempt to obtain an effect on a
photographic plate that could be ascribed to a mitogenetic radiation.
MITOGENETIC RADIATION 287
In the first experiments an effort was made to detect photographically
an emission of M-rays from growing onion roots. The technical dif-
ficulties were considerable for the roots had to be kept moist while the
photographic plate had to be kept absolutely dry to avoid the possibility
of the so-called Russell effect.1 Furthermore, while maintaining these
conditions, the plate had to remain within a few centimeters of the roots
for a considerable length of time.
To surmount these difficulties we placed a photographic film (East-
man Kodak Superspeed) inside of a quartz flask, which was blown espe-
cially for this work with a flat bottom only 0.2 mm. in thickness,2 cer-
tainly thin enough not to offer any serious resistance to an ultra-violet
radiation of 2000-2300 A°. While Schumann plate or film is more
sensitive in the 2000 A° region (five times as sensitive to the 1850 line
according to Adam Hilger, Ltd.) than ordinary film, Schumann film is
notoriously unstable and it is doubtful if it could be used with such long
exposures as are described later. Spectrograms of the Al spark taken
with a Hilger quartz spectrograph on Kodak superspeed film (5 seconds)
show the 1990, 1935, 1862 and 1854 lines although the last two are faint.
Quartz test-tubes of 1.8 mm. wall thickness passed all the above lines,
although there was undoubtedly considerable absorption of the last two
but practically no absorption with 0.2 mm. thick quartz.
The film was pressed gently against the bottom, emulsion side down,
with cotton or glass wool packing. The exposure was made by invert-
ing the flask beneath the growing roots of an onion, the roots being al-
lowed to grow down toward the flask from a distance of approximately
three centimeters until they touched the upturned bottom of the flask,
one half of which had been covered with a cover-glass to serve as con-
trol.3 Negatives were exposed in this way to growing onion roots for
2]/2, \2l/2, and 24 hours. When developed, they showed no effects of a
radiation of any sort.
These results are quite in keeping with those of other workers who
have tried exposures of 48 hours' duration. To increase this exposure-
time yeast cultures were substituted for the onion roots, Fleishman's
yeast being used, and no attempt made to keep the culture pure.
1 The effect which various substances such as metals, cod liver oil, gelatin,
gutta percha, celluloid, collodion, and certain vapors, etc., may exert upon a
photographic plate through their oxygen-absorbing capacity. Hydrogen peroxide,
one part in one million of water, will influence a photographic plate in eighteen
hours. See Russell, W. J. (Proc. Roy. Soc., London, 1899, 64: 409-419, and Pho-
tograph. Jour., 1899, 23: 91-97), and Kugelmass and McQuarrie (Science, 1925,
62: 87-88).
- Engineers of the General Electric Company estimate that quartz of this
thickness will transmit 75 per cent of the ultra-violet light of 2000 angstroms wave
length striking it. (Personal communication.)
3 Which Rawin (1924) has found to be impenetrable to the M-rays.
288 G. W. TAYLOR AND E. X. HARVEY
some «it" the experiments with yeast as a source of radiation, the
quartz flask described above was used, the yeast culture being placed in
the flask, and the flask being placed over a piece of cut film, one half of
tin- area under the flask being covered with black paper to serve as a con-
tn>l. Care was always taken that the flask should not press directly
against that area of the negative being exposed.
Exposures were made in this way for 13, 100, and 144 hours without
affecting the exposed negative in any way. Microscopic examination of
the yeast culture following, and during, each of the above exposures
Allowed that in the medium used, Pasteur's with sugar, the budding
activities of the yeast had just about stopped at the end of the five days,
but that up to that time-, budding plants in the culture were very nu-
merous.
To obtain a still longer exposure to an actively budding yeast colony,
the flask containing the culture was arranged as before over a negative,
but this time the culture medium in the flask was renewed at intervals of
from 2-3 days. This was done by merely pouring out the old medium
and pouring in the new. enough yeast adhering to the flask to insure the
proper inoculation of the new medium. After the solution had been
changed the flask was returned to its original position over the negative,
one half of the negative being covered to serve as a control to the other
half. This experiment was set up in absolute darkness and remained in
darkness for the whole length of the exposure, including the times when
the change of culture medium was being effected.
In this way a negative was obtained that had been exposed to a con-
stantly fresh and actively budding yeast colony for 15 days (360 hours).
but even after this long an exposure the- negative on being developed
showed absolutely no effect from a radiation of any sort, in spite of the
fact that an onion root exposed to such a culture for 20 minutes is sup-
posed to show a marked mitogenetic effect from the exposure. Frank
(1929) states that the onion root is 600 times more sensitive than the
photographic plate. If an onion root can be affected in 20 minutes, a
plate should be affected in approximately 600 X 20 minutes or 200 hours.
In this experiment the plate was exposed for 360 hours.
The third series of experiments was set up in the following way. A
series of quartz ( 1.1 to l.S mm. thick) and glass test-tubes, each con-
taining a >trip of cut film and each tightly stoppered to exclude all
vapors, were partly immersed in bowls containing yeast culture. The
upper part of each tube extended above the yeast culture so that in addi-
tion to tin- control negatives contained in the glass tubes, the upper part
of each negative in tin <|iiart/ tubes served as a control to the lower por-
tion which extended into the yeast culture. The change of culture
MITOGENETIC RADIATION 289
medium, made at intervals of 2-5 days, was easily and quickly accom-
plished by merely removing the tubes from the bowls containing the old
medium and placing them in bowls containing the new. This series was
set up and remained in complete darkness for the whole time the ex-
posure was being made.
The negatives from the quartz tubes with their controls from the
glass tubes were removed and developed simultaneously at intervals of
10, 40, 52, 80, and 89 days. But the negatives showed no effects from
these long exposures to a supposedly potent source of mitogenetic radia-
tion. A portion of each film used was exposed to light, and on develop-
ment showed normal blacking, indicating that they were in good con-
dition.
DISCUSSION
If onion roots and yeast gave off a radiation of the nature of ultra-
violet of an intensity sufficient to exert the potent physiological effects
attributed to it, we believe that in 89 clays it should have affected in a
noticeable manner a photographic plate exposed to them. In view of
the fact that such an effect was not obtainable, we cannot help but con-
clude that these tissues do not give off such a radiation, and that if the
tissues exert a mitogenetic effect it must be through some agency other
than the emission of ultra-violet waves.
These experiments do not, of course, exclude the possibility that
exposure of cells to minimal amounts of ultra-violet light may stimulate
cell division. They make it practically certain, however, that budding
yeast and onion root tips produce no ultra-violet radiation of 2000 A°
or longer. Indeed, the emission of such a radiation from a living cell
would be extremely unlikely. Even luminous animals, which produce
visible radiation, have a spectrum which stops far short of the ultra-
violet. Experiments with Cypridina luminescence whose spectral
maximum lies at A =^A8/j. (Coblentz, 1926) show that this light pro-
duces exactly the same effect on a photographic plate when exposed
through quartz as when exposed through glass. No difference in den-
sity after development could be detected between the quartz- and glass-
protected regions from exposures not long enough to give maximum
blackening, i.e., care was taken to make the exposure correspond to the
region of the plate where increased exposure gives increased blackening.4
No chemical reactions in aqueous media have ever been definitely
shown to emit ultra-violet light. The alleged effects of this kind
(Matuschek and Nenning, 1912) have been found to be due to the action
of vapors (Mathews and Dewey, 1913), an effect against which too
great precautions cannot be taken.
4 Harvey, E. N., unpublished experiments carried out in 1925.
20
290 G. W. TAYLOR AND E. X. HARVEY
\\Y believe that a root-tip, in which cell division is observed to be far
from uniform, is unfortunate material to work with. Those who have
studied yeast know the difficulties of estimating budding under different
conditions. Indeed, if the division or budding of one cell can affect the
division, or budding, of another, we should find the growth of organisms
•ncrease more rapidly than corresponds to logarithmic increase. Rate
of cell division should be dependent on volume of suspended organisms.
\*<i such effect has been recognized. The allelocatalytic phenomenon
of Robertson has not been gnu-rally confirmed by other workers.
(Richards, O. W., Thesis.) The alleged mitogenetic effect could be due
to insufficient control of experimental material, primarily difficult to
work with. For this reason and in view of the alleged effects from
material containing no dividing cells, hemolysed blood, contracting
muscle, conducting nerves, etc., we are inclined to place them in the same
category as the famous n-ray r> of Hhmdlot (1903), shown to be a purely
subjective phenomenon by Wood (1904) and (iehrcke (1905).
CONCLUSIONS
1. Evidence is offered to show that the onion root as a detector otf
the mitogenetic rays cannot be relied on, since in normal roots, unexposed
to any supposed source of mitogenetic radiation, there may still be a
variation in the number of dividing cells in the two halves of a root as
high as 50 per cent.
2. Exposure of a photographic plate to growing onion roots through
.2 mm. quart/ fur 4S hours failed to detect a mitogenetic radiation.
3. Exposure of photographic negatives to an actively growing and
dividing yeast culture through quart/ 1.1 mm. thick for as long as 89
days failed to affect the negative in any way.
4. The authors conclude in view of their negative evidence that the
existence of a mitogenetic radiation in the form of ultra-violet light by
normally growing onion roots and yeast plants cannot be accepted as a
fact.
BIBLIOGRAPHY
ANIKEN, A. W., 1926. Das Xervensystem als Quellc mitogenetischer Strahlung.
.Irch. /. entw.-Organism., 108: 609.
BAKON, M. A., 1926. tvber mitogenetische Stralilnnj; ln-i Protisten. Arch. f.
entw.-Organism., 108: 619.
BARON, M. .V, 1('_'X. P>aktericn als Quellen mitogenetischer (ultravioletter)
Str;ililung. Ccntralbl. /. Rakl., Parasit. n. Injckt., Aht. 2, 73: 373.
BI.ACHER, I I , 1'i.W. Die Rolle der mitogenetischen Strahlungen in den Prozessen
der Metanmrpliase der Schwanzlosen Amphibian. Arch. f. entw.-Orga-
nism., 122: 48.
•"' For further literature on this subject see Charpentier, Lambert and Meyers
in the Comfit, rend. Soc. Biol. (Paris) for 1904-05.
MITOGENETIC RADIATION 291
BLACKER, L. J., AND N. W. BROMLEY, 1930. Mitogenetische Ausstrahlungen bei
dor Schwanzregeneration der Urodelen. Arch. /. entw.-Organism., 123:
240.
BLACKER, L. J., AND O. G. HOLZMANN, 1930. Resorptionsprozesse als Quelle der
Formbildung. III. Mitogenetische Ausstrahlungen wahrend der meta-
morphose hei Urodela. Arch. f. entw. -Organism., 123: 230.
BLONDLOT, R., 1903. Sur la propriete d'emettre des rayons n, que la compression
confere a certains corps, et sur remission spontanee et indefinie de rayons
n par 1'acier trempe, le verre trempe, et d'autres corps en etat d'equilibre
moleculaire contraint. Coinpt. rend. Acad. Sci., 137: 962.
BORODIN, D. N., 1930. Energy Emanation during Cell Division Processes (Mi-
Rays). Plant Physio!.. 5: 119.
BROMLEY, N. W., 1930. Der Einfluss der primaren Vcrheilung der Wunde auf
die Entstehung mitogenetischer Ausstrahlungen in ihr. Arch. f. cntiv.-
Oryanism., 123: 274.
CHOUCROUN, N., 1930. On the Hypothesis of Mitogenetic Radiation. Jour. Mar.
Biol. Assn., Plymouth, 17: 65.
COBLENTZ, W. W., AND C. W. HUGHES, 1926. Spectral Energy Distribution of the
Light Emitted by Plants and Animals. Bur. Stand. Sci. Papers, 21: 521.
FRANK, G., 1929. Das mitogenetische Reizminimum und -maximum und die Wel-
lenlange mitogenetischer Strahlen. Biol. Zentralbl., V49: 129.
FRANK, G. M., AND A. GURWITSCH, 1927. Zur Frage der Identitat mitogenetischer
und ultravioletter Strahlen. Arch. j. entw.-Organism., 109: 451.
FRANK, G., AND M. KUREPINA, 1930. Die gegenseitige Beeinflussung der Seeige-
leier als mitogenetischer Effekt betrachtet. Arch. j. entw.-Organism.,
121: 634.
FRANK, G., AND M. POPOFF, 1929. Le rayonnement mitogenetique du muscle en
contraction. Com fit. rend. Acad. Sci., 188: 1010.
FRANK, G., AND S. SALKIND, 1926. Die Quellen der mitogenetischen Strahlung
im Pflanzenkeimling. Arch. f. entiv.-Oryanism., 108: 596.
FRANK, G., AND S. SALKIND, 1927. Die mitogenetische Strahlung der Seeigeleier.
Arch. f. entw.-Organism., 110: 626.
GEHRCKE, E., 1905. Zur Deutung der Versucke mit N-Strahlen. Pltysik.
Zeitschr., 6: 7.
GESENIUS, H., 1930. Uber die Gurwitschstrahlung menschlichen Blutes und ihre
Bedeutung fiir die Carcinomdiagnostik. Biochcm. Zeitschr., 226: 257.
GURWITSCH, A., 1923. Die Natur des spezifischen Erregers der Zellteilung. Arch.
f. cnt^v. Mcch., 100: 11.
GURWITSCH, A., 1924. Les problemes de la mitose et les rayons mitogenetiques.
Bull. d. Histol. Appl., 1: 486.
GURWITSCH, A., 1926. Das Problem der Zellteilung physiologisch betrachtet.
Monogr. Gesamtgeb. Physiol. Pflanz. u. Tiere, No. 2. Springer, Berlin.
GURWITSCH, A., 1928a. Some Problems of Mitogenetic Rays. (Russian paper
with German summary.) Jour. Soc. Bot. Rnssic Acad. Sci., 13: 179.
GURWITSCH, A., 192S/>. Einige Bemerkungen zur vorangehenden Arbeit von
Herrn B. Rcssmann. Arch. j. entw.-Organism., 113: 406.
GURWITSCH, A., 1929. Uber den derzeitigen Stand des Problems der mitogenet-
ischen Strahlung. Protoplasma, 6: 449.
GURWITSCH, A., AND L. GURWITSCH, 1925o. Weitere Untersuchungen uber mito-
genetische Strahlungen. Arch. f. mikr. Anat., 104: 109.
GURWITSCH, A., AND L. GURWITSCH, I925b. Uber den Ursprung der mitogenet-
ischen Strahlen. Arch. f. entw.-Organism., 105: 470.
GURWITSCH, A., AND L. GURWITSCH, 1926. Die Produktion mitogener Stoffe im
erwachsenen Tierischen Organismus. Arch. f. entw.-Organism., 107: 829.
292 G. W. TAYLOR AND E. X. HARVEY
GURWITSC H, A.. AND L. GURWITSCH, 1928. fber ultraviolette Chemolumineszenz
der Zellen im Zusammenhang mit dem Problem cles Carcinoms. Biochem.
Zeitschr., 196: 257.
G i. A.. AND N. GURWITSCH, 1924. Fortegesetzte Untersuchungen iiber
mitogenctische Strahlung und Induktion. Arch. f. entw.-Organism., 103:
68.
G rsCH, I,.. AND A. AXIKIX, 1928. Das Cornealepithel als Detcktor und
Sender mitogenetischer Strahlung. Arch. f. entw.-Organism., 113: 731.
WlTSCHj A., AND G. FRAXCK. Sur les rayons mitogenetiques et lour identite
avec les rayons ultraviolets. Compl. rend. Acad. Scl., 184: 903.
Gi'TTEXBERG, H. vox, 1'L'Si/. Die Theorie der mitogenetischen Strahlen. Biol.
/.cntralH.. 48: 31.
GUTTEXBERG, H. vox, ln28/>. Schlusswort v.ur Arbeit von B. Rossmann. Arch. f.
entw.-Organism., 113: 414.
HOLLAEXDKK. A., AND 1'". ScHOEFKEi., 1931. Mitogenetic Rays. Quart. Rcr. Biol.,
6: 215.
HOLZMAXX. O. G., 1930. Mito^enetischc Ausstrahlungen \vahrend der Meta-
morphose bei Drosopbila melanogaster. Arch. f. entw.-Organism., 123:
266.
Kisr.iAK-STKATKK.wiTscM. M., 1"27. Das mitogcnetisclie Strahlungsvermogen des
Kartoffelleptoms. Arch. f. entw.-Organism., 109: 283.
M Ai.unr. I.. AND M. MAI.ROU, 1927. Recherches sur les radiations mitogenetiques.
Bull. d'Histol. App}.. 4: 253.
M AU'i'i . I.. AMI M. A! \(,Knr, In28. Action a distance du Bacterium tumefaciens
sur le developpenu-nt de Foeuf d'Oursin. Conipt. rend. Acad. Scl., 186:
802.
MAGROU, ].. M. MAGROU, AND F. CIIOUCROUX. 1929. Action a distance du Bac-
terium tumefaciens sur le developpement de 1'oeuf d'Oursin (nouvelles
ii ricnces). Coinpt. rend. .lead. Sci.. 188: 733.
MATMKWS. ]. II.. AND L. H. DIAVKV, 1913. The Production of Photochemically
Active Kays in Ordinary Chemical Reactions. Jour. Plivs. Cliein., 17:
230.
MATT-I in K AND XKXXIXG, 1912. Ueber das Auftreten von cliemisch wirksamen
Strablung bei chemischen Reaktiourii. Chein. /.ellunii. 36: 21.
POTOZKY. A., AND Q. Zoi.i.iXK, 1929. Untersuchunsifii uber die mitoneiu'ti>che
^iralilun.L; des Blutes. Hioeliein. /.citsehr.. 211: 352.
R \\VIN, \Y., 1^24. \\'eitere r,i-ilra»e 7.\\r Kenntnis der mitotischen Ausstrablung
und Induktion. Arcli. /. inikr. .-liuit.. 101: 53.
Rii'iiK. T.. TND I). (i.uioK. l'*2S. Xellteilung und Strahlung. Springer, I'.erlin.
KOSSMAXX, B., 1928. Untcrsucluingen iiher die Theorie der mitogenetischen
Strahlen. Arch. /. entw.-Organism., 113: 346.
ROSSMANN. B., 1928-29. Mitogenetischc Inductionsversuche mit Ilefe aU Indi-
i-ator. ./;•<•/;. /. entw. -Organism., 114: 583.
KI . P. G., 1925. \Yeitere untersuchungcn iiher mitogeiu-tiscbe Strahlen und
Induktion. Arch. f. vrikr. Ann!.. 104: 121.
SALKIND, S., VH{). f'lier den Rhythmus der mitoiieni-ti^clien Strahlung bei der
I'.ntwicklung des SeeigeK-ies. Arch. /. entw.-Organism., 115: 360.
iKi.ND. P(no/.KY. i MI /cci.iNA, l'^3(i. hie in ito.ueiiet isd le IVeinflussung der
l.ier von Protodrilus und Saccocirrus. Arch. /. entw.-Organism., 121:
630.
}•/. WALTER, l''2S. Das Problem der mitoLM-netisi-hen Strahlen. Biol.
Zcntralbl, 48: 302.
. \\'. \\"., l('2Sii. C'ber (lie mitogrnetische Strahlung des Arbeitsmuskels
und t ini'.M r ainlerrr Gcwebc. li'wchcm. Zeitschr., 202: 115.
. \\". \\'., l'>2X/i. flier die t'rsachen der mitogenetischen Strahlung. Bio-
cliein. Zellschr., 202: 123.
MITOGENETIC RADIATION 293
SIEBERT, W. W., 1930. Die mitogenetische Strahlung cles Blutes und des Harns
gesundcr und kranker Menschen. Biochcin. Zcitschr., 226: 253.
SORIN, A., 1926. Zur Analyse der mitogenetischen Induction dcs I.lutes. Arch.
f. cntzi'.-Organisin., 108: 634.
SORIN, A. N., 1928. liber mitogenetische Induction in fruheren Entwichlungs-
stadien des Huhnerembryo. Arch. /. cnlu'.-()r<ianisin., 113: 724.
STEMPELL, W., 1929. Die Lebensstrahlen. Strahlcnthcniphic, 34: 868.
URBANOWICZ, K., 1927. Gurwitsch's mitogenetische Strahlung, an Paramazien-
teilungen Gepri'ift. Arch. f. cntiv.-Organism., 110: 417.
WAGNER, N., 1927. Uber den von A. Gurwitsch entdeckten spezifischen Erreger
der Zellteilung (Mitogenetische Strahlen). Biol. Zcntralbl., 47: 670.
WASSILIEW, L. L., G. M. FRANK, UND E. E. GOLDENBERG, 1931. Versuche iiber
die mitogenetische Strahlung des Nerven. Biol. Zcnlnilhl., 51: 225.
WOOD, R. W., 1904. Die N-Strahlen. Physik. Zcitschr., 5: 789.
SOME OBSERVATIONS ON THE EGGS OF FUCUS AND
II '< ).\ Till '.IK MUTUAL INFLUENCE IN THE DETER-
MINATION OF THE DEVELOPMENTAL AXIS
D. M. WHITAKER 1
i /•><>»( I lie Laboratory of General Physiology, Harvard University,
Cambridge, Mass.)
Several species of the seaweed Fucus arr obtainable in abundance on
the rocky shores of Nahant peninsula, outside the entrance to Boston
Harbor. These sea \veeds have a long breeding season, shedding eggs
and sperm in the winter months when other types of eggs are difficult to
obtain. In 1929-30 eggs of Fucus vesiculosus were collected from Oc-
tober until June. Throughout this time some eggs were always ob-
tainable from plants with relatively large fruiting tips, or receptacles,
and in February. March. April, and May they were obtainable in great
abundance.
Fucus vcsiculosus is dioecious. The sexes can be sq^arated by sight
with fair reliability if a cut receptacle is examined. The conceptacles
of the male plants are orange due to the carotinoids contained in the
antherozoid or sperm cells. The conceptacles of the female plants are
green or brownish-L;reen, largely because of the plastids in the eggs.
Identification becomes certain if a thin section of the receptacle is ex-
amined microscopically.
The purpose of this paper is to present a number of observations and
experiments, some of which have been incidental to measurements on
respiration in the Fucus eggs which will be presented elsewhere. The
results recorded here have particularly to do with the nature of the eggs
and with certain factors involved in determining the first division plane.
The first division of the Fucus egg ordinarily -ivrs rise to two cells of
different shape. One, which includes the rhyxoidal protuberance, is the
parent cell for the formation of the rhy/oid, the other gives rise by di-
visions to the thallus. At the first division, therefore, the polarity or
developmental axis of the spore has been determined ami is first indi-
cated. A number of environmental factors have been found capable of
determining the cleavage plane and tin- polarity of the Fitcus spore.
Orientation of the cleavage by a directed beam of light has been demon-
strated in a number of plants (e.g., Pierce, 1906). Farmer and Wil-
liams i IS' i,X) have shown that if fertilized Fucus eggs are illuminated
National Research Council Fellow in the Biological Sciences.
294
EGGS OF FUCUS AND DEVELOPMENTAL AXIS 295
from one side, the rhyzoids usually originate on the side of the egg
remote from the light. Miss Hurd (1920) found that in F. inflatus (a
monoecious species, collected in San Francisco Bay) when directed
beams of different light frequencies are used, red light has no effect but
the short blue orients the cleavage and the direction of growth of the
rhyzoid.
At Friday Harbor Lund (1923) passed an electric current through
sea water containing spores of the monoecious FHCHS inflatus. Cross
streams of sea water were designed to carry off electrolytic products
formed at the electrodes. A potential drop of 25 millivolts across the
diameter of an egg oriented the division plane and the developmental
axis. The rhyzoid cell came to lie toward the -(- pole.
There are no visible marks or identifications of polarity in the un-
fertilized FUCKS egg. The nucleus lies in the middle of the cell. When
spores which are not too greatly crowded develop in the dark, the direc-
tions of divisions lie entirely at random. Miss Hurd (1920) observed
in her work with colored light that if the eggs lie close together, within 2
or 3 egg diameters, they tend to send out the rhyzoid toward each other, or
toward the center of a nearby mass of eggs. This phenomenon, which
she calls a " group effect," was especially pronounced in the dark, but
was strong enough to overcome the orienting effect of the directed light
if the eggs were close together.
This directive effect of one egg on another at a distance presents
some points of interest. It might possibly be due to a differential of
oxygen tension, or of C(.)2, or to the accumulation of some other me-
tabolite. If the cells are either giving off or consuming some substances
in radial fashion, in the case of neighbors the additive effect on the inter-
vening space between them would cause that part of the sea water en-
vironment to be most altered. No jelly or solid substance of the eggs
traverses this space. YVinkler (1900) attempted to establish a gradient
of oxygen tension across the spores of Cystosira barbata, which are also
oriented by light, to see if this might be the determining factor. The
results were negative. In view of recent work purporting to show an
effect at a distance upon dividing cells due to " mitogenetic " radiation,
the possibility of some such effect in the " group effect ' in Fucus
presents itself. Mitogenetic rays have been supposed to affect primarily
the division rate of neighboring cells, rather than the polarity or plane
of division. It is possible that this qualitative distinction is not justified,
however, as the plane of division may be determined by asymmetric or
differential rates of the processes leading to cell division. The experi-
ments which are to be described do not discover the nature of this in-
fluence of cells at a distance. It is at present possible to answer the
296 D. M. WHITAKER
following two questions: 1. In order that cells shall exert this influence
upon a neighbor, must there be nuclear activity or cell division in the
directing cells? _'. Is the directive effect in FUCKS sped tic? The an-
swers to these questions still leave open the question of a possible role
of mitogenetic ray>. since this type of radiation has been as well demon-
strated coming from non-dividing tissues, even recently macerated tis-
sue. a> trom dividing cells. A brief review of some of this work is
given by llullaeiider and Schocffel. 1'Ml. J;urther experiments to con-
tinue these preliminary observations are planned.
lief on- proceeding to a consideration of these experiments, howe\er,
a description of the Fnens eggs, and the results of some other observa-
tion-, will be presented.
The (.iainetes of Fncits
The I'ucus plant is diploid. As in animals the haploid generation is
confined to the gamete-, Yamanouchi ( 1( )()',)) has estimated the
chromosome number in /•". -fcsicnlosits to be ()4-32. After the plants had
been brought into the laboratory the sex of each plant wa> determined
by microscopical examination of a section of a receptacle. The re-
ceptacle- we're then cut off and placed in covered glass dishes in an ice
box at about 3° ( '. 1 '-nally within 24 hours or less the gametes within
their capsules had been extruded from the' conceptacles to lie in mounds
on the outside of the receptacles. Removal of the plants from sea water
induce- -bedding. In nature the -bedding is stimulated when the plants
are stranded at low tide. Too much drying, however, is higlilv detri-
mental, and it is hot to keep the receptacles in covered dishes as well as
to keep them cool.
The eggs lie S in a capsule whrn shed. The capsule wall consists
of two thin membranes, probably with a gelatinous substance between,
ID microns wide. When Kl -f L is added, and then H_.S< ^, the capsule
membranes and the space between them do not turn blue (although the
eggs within do ) . Thi.s indicate- that the capsule is not made of cellulose.
When MX' '; and XII. are added these' membrane-, and especially the
space between them, turn deep yellow-brown, indicating that they are
proteinaceous. The membranes coagulate and partlv dissolve when the
acid i- added, and some coagulation of tin- substance between the mem-
branes is observed. More slowly the egg- also turn yellow, although
the color of the plastid- tends to some extent to obscure the color.
When tin egg capsules are washed off from the fruiting tips into
normal sea water, after a time, depending on temperature, first the outer
and later tin- inner membrane of the capsule breaks and the eggs are
aborted into the sea water, often being somewhat squeezed together in
EGGS OF FUCUS AND DEVELOPMENTAL AXIS 297
passing out. If the eggs arc kept cold, they apparently undergo no
deterioration for many clays. High percentages of fertilization and of
normal spores have been obtained from eggs which had remained un-
fertilized, either in light or dark, for more than a week.
Many diflagellate sperm or antherozoids occur in small capsules (64
antherozoids per capsule, according to Yamanouchi, 1000). These cap-
sules are exceedingly delicate and appear to dissolve completely in the
sea water. The sperm are immotile when liberated, but they begin to
move in a few seconds, and are soon swimming actively. My own
experience has been that their life is short, or rather that they do not
fertilize eggs well after a few hours, especially if they have been in
dilute suspension. They may be kept a number of days, however, if
they remain dry on the fruiting tips in a moist chamber. They are im-
motile in this condition. Sperm suspensions of even moderate concen-
tration are brilliant orange.
The unfertilized eggs of FUCKS trsicitlosiis vary considerably in size.
Single eggs varied between 52 and 70 microns in diameter, averaging
between 60 and 65. Larger eggs are found which will be discussed later.
The eggs are readily stratified by centrifuging, and develop normally
after stratification. Eggs centrifuged immediately after fertilization
tend to become amoeboid but eventually round up. Centrifuging for 20
minutes at 2600 r.p.m. (20 cm. radius, 18° C.) throws almost all of the
formed bodies to one end. A gray cap occupies the end position. Its
thickness is about one ninth the diameter of the egg. The nucleus lies
at its inner edge, just between it and the plastids, which are densely
crowded into a zone bordering the gray cap. A few plastids remain
behind adhering to the peripheral regions of the egg. The stratified
materials are all less dense than the cytoplasm at large, as the gray cap
is seen to float uppermost when the eggs settle in a tube and are ob-
served from the side as they fall. The eggs are comparatively dense.
They settle more rapidly than the eggs of such animals as Arbacia,
Cumingia, Chostopterus, etc., in spite of their smaller size. The vol-
ume of materials which are moved by the centrifuge is much less in
proportion than in such animal eggs as the sea urchin's. Since the eggs
are photosynthetic they are perhaps able to dispense with as large a store
of food materials.
The unfertilized eggs may readily be cut with the microneedle, al-
though unless they are pinched gently with care, they burst and disperse
their substance into the sea water. They appear to be only poorly pre-
pared to gelate a new surface on an exposed cut, being deficient in ma-
terials for what Heilbrunn has called the " surface precipitation reac-
tion."
\
298 D. M. WHITAKER
Fertilisation
When a heavy sperm suspension is added to unfertilized eggs, the
rs arc rapidly rotated by the sperm. After a few minutes they stop
rotating even though the sperm are still fully active. That this whirling
of the eggs has no significance in bringing about fertilization is evident
Cl) because dilute sperm suspensions which fertilize the eggs cause no
rotation, and (2) because concentrated sperm of another member of the
Fitcaccff, Ascoph\lhnn nodosum, which do not fertilize the eggs of
Fncus rcslculosHs rotate the Fitcus eggs more rapidly than the eggs of
their own species (which are larger ).
Fertilized eggs were placed in a thermostat at 18° C.. and the time-
lapse to the first division was noted. The time span over which first
divisions occur in a population is a wide one. covering approximately
U to 18 hours. Fifty per cent of the eggs were observed to have di-
vided after 14 hours in one case, and after 15 hours in another. At
cold temperatures the time is greatly extended.
The Fitcits egg has been described as secreting a cellulose wall im-
mediately after fertilization. Eggs were fertilized and then cut with the
microneedle at increasing intervals after fertilization, to determine the
physical nature of this secretion and the sequence of physical change.
It has been stated that the unfertilized eggs may readily be cut in half,
although they burst very easily. When they burst no membrane of any
-ort is visible around the egg or remaining behind. Five minutes after
fertilization the eggs may be cut with the greatest ease. There is no
longer any tendency to burst. When the fragments are separated with
the needle a >ticky gelatinous material, which has been secreted, is seen
to bridge across between the fragments. At 10 minutes conditions are
about the same. Fifteen minutes after fertilization a slight increase in
the rigidity of this gelatinous secretion is noticeable. After 30 minutes
it is slightly more firm. At 45 minutes it lias become a definite semi-
solid wall which holds its shape but which is still easily cut. After an
hour the Accretion has become so linn as to be cut only with great dif-
ficulty, and usually only after bursting the egg within. A fur an hour
and a half the wall i> tough and rigid. The eggs dodge the needle and
cannot be cut.
It is probably this sticky secretion which cause- the eggs to adhere
to the substrate and stop rotating after fertilization. The eggs adhere
to the substrate with increasing tenacity until by _" .'• hours they are fairly
well attached, or if they are kept in suspension by gentle shaking, after
2y2 hours they are elumped together in adhering masses. After the
secreted wall has become firm the eggs are well protected and may at any
EGGS OF FUCUS AND DEVELOPMENTAL AXIS 299
time be dislodged without damage from the substrate to which they
adhere.
The secreted wall of the fertilized egg was subjected to colorimetric
cellulose and protein tests. The walls of 24-hour spores gave no protein
reaction, although of course the egg itself and cytolytic extrusions from
it turned deep yellow. The cell walls turned brilliant blue when sub-
jected to Kl -|- l._, -\- H2SO4. a test for cellulose. No rotation of polar-
ized light was observed with a polarizing microscope, but this may well
have been because the cellulose is too thin to cause a detectable amount
of rotation.
The unfertilized eggs show a faint blue color in the cellulose test,
perhaps due to the material in the cortex which is to be secreted. Ten
minutes after fertilization the color reaction of the secreted jelly is
equally faint. At 25 and 55 minutes there is a slight increase in the blue
reaction. At about an hour and a half the first decidedly blue color
results. Even this, however, is not as brilliant a blue as in the 24-hour
eggs. It seems probable that a relative of cellulose is secreted as the
sticky jelly, and that gradually the crystal molecular arrangement is
assumed so that by an hour and a half the tough cellulose wall has been
formed, which gives the blue color reaction. As the rhyzoid grows out
on one side this membrane must be softened and added to; at any rate
it precedes and sheathes the early rhyzoid. Nevertheless it gives a
brilliant blue color reaction at the tip as if it does not reconvert back into
the faintly coloring jelly.
If eggs which have been fertilized for an hour and a half are placed
in sea water saturated with dextrose they shrink and collapse. Un-
fertilized eggs do not collapse. Eggs which have been fertilized for 25
minutes do not collapse, while those which have been fertilized for 45
minutes collapse to some extent. Eggs which have been fertilized for 3
hours or longer collapse and leave behind the transparent cellulose wall,
which is clearly revealed in this way. Farmer and Williams note that
placing the eggs in tap water so that they burst and flow out also re-
veals the cell wall. The difference in behavior in dextrose sea water is
often a convenient and fairly reliable way of testing for fertilization, as
the jelly and the cellulose wall are not themselves readily seen in the
normal fertilized egg. After two and a half hours in dextrose-saturated
sea water the eggs have mostly rounded out and recovered their shape.
Shrinking with sugar in the later stages causes the individual cells of the
spore to stand out clearly. Even the early spores have a remarkable
ability to withstand dehydration and to develop normally after being
returned to normal sea water. Thus embryos one hour, an hour and a
half, and two hours after fertilization were placed in dextrose-saturated
300 D. M. WHITAKER
sea water for two hours. They were then returned to normal sea water,
and developed in typical fashion.
The Origin and Fate of Giant Eggs
The sixe <>f single eggs of Fncns vesiculosis eggs varies considerably
(about 52-70 microns). In addition, a number of giant eggs, often
much larger, are frequently found in a sample. The proportion ()f these
giants depends greatly on the treatment to which the eggs have been
subjected before their emergence from the capsules. In some cases
more than half will be giants, most of which are much larger than the
<-ggs seen within the capsule. This difference in sixe led IVhrens, iSSd
(cited by Farmer and Williams, 18'H>), to propose that the large eggs
represent a stage in fertili/ation. Farmer and Williams (lS"n. 1X(>8)
point out that by no means are all fertili/ed eggs larger than the un-
fertili/ed eggs in tin- capsules, and they further noted that two or three
nuclei are sometimes visible in these large eggs, which they therefore
regarded as abnormal eggs. Fxamples of what appear to be giant eggs
appear in photographs by llurd ( 1()20) and Lund (1923). It might
be supposed that these large multinucleate eggs result from the failure
of certain of the parent germ cells to divide. While this may possibly
be the case sometimes, careful direct observations have shown another
and simple origin of frequent occurrence.
When the eggs eoine out of tin- capsule they are often pressed to-
gether. At this time they often fuse' to form giants. These giants
have been observed to form as the result of the fusion of 2. 3, 4, 5, 6. 7,
and 8 single eggs. The giant egg. of course, contains the corresponding
number of nuclei, although possibly these fuse later. Once eggs have
fused, they have- ne\er been seen to separate later. The effect of tem-
perature at the time of emergence is very marked. Samples of the same
set of capsules were divided into lots, some of which were placed at
3° C. and some at between 25° C. and 28° C., during the period of
break-down of the capsules and release of the eggs. At the lower
temperature the break-down of the capsules takes a much longer time,
but after the eggs were out this striking difference was found: In the
.•s at 3' ( '. only 3 giants were found in about f >()<>() eggs. In the eggs
which had emerged at 25-28° C. between 50 per cent and (>() per cent of
all the eggs were giants, mostly of about 4 fused eggs. The high tem-
perature apparently alters what might loosely be called the consistency
,,i tl|,. eggs, -1' that thev are much more prom- to fuse. One of the
benefits ot kee] ing the eggs eold while they are emerging from the cap-
sules is to a\oid these giants in experiments in which they are to be
EGGS OF FUCUS AND DEVELOPMENTAL AXIS 301
avoided. The plants should be kept cold from the time of collecting, as
otherwise some fusions will take place within the capsules.
The development of these giants involves some peculiarities and
variations. No doubt they occur to some extent in nature when the eggs
are shed on warm days. Many hundreds of individual eggs were
isolated and fertilized, and single mononucleate eggs were never ob-
served to undergo abnormal cell division nor to give rise to more than
one rhyzoid (although this in some cases branched at a later time).
The giants also usually divided in normal fashion, with one giant rhyzoid,
but not infrequently two and sometimes three independent rhyzoids
developed simultaneously from various parts of the egg. The number
of rhyzoids bore no special relation to the number of cells which had
fused to form the giant egg, except that there were never more rhyzoids
than component cells. Thus, isolated giants which had been seen to
originate from the fusion of six eggs (for example) formed in some
cases one, in others two, and occasionally three rhyzoids. Most com-
monly when two rhyzoids formed they grew out side by side, sometimes
having a common basal part. Not infrequently, however, when reared
in the dark and away from neighbors, two rhyzoids grew out 180° or
less apart. A few examples are sketched in Fig. 1.
FIG. 1. Spores from fused eggs (F. vcsiculosis). The small left-hand spore
is a single egg.
Sometimes one of the supernumerary rhyzoids, especially if it is
small, may disintegrate and sluff off, while the remainder of the spore
then develops normally.
The causes which determine whether twinning, or what degree of
twinning, shall take place in these giant spores are not known. Two
possibilities suggest themselves. It may depend upon how many of the
constituent egg cells receive sperm independently, or it might depend
upon the degree of fusion of the constituent cells, as determining whether
the egg nuclei may come together to form a single polyploid nucleus,
or whether internal partitions (former cell surfaces) persist, holding
the nuclei apart so that they become independent division-centers. At
any rate, the relation between the twinning and the fused origin of the
302 D. M. WHITAKER
giant e.^s indicates that the phenomenon is of the same general sort as
the clu].liY;iti"ii in parts which results from experimentally fused eggs
or hlu-ml;e (if sea urchins and starfishes (cf. Morgan's text, 1928).
Cross-Fertilization
In the spring gametes were ohtaincd from the monoecious Fitciis
r: < -iicscciis.2 The eggs and sperm are shed simultaneously in a heavy
mass of vise-mis jelly which covers the entire fruiting tip. These eggs
are larger than those of F. vesicnlosns, averaging about 85-90 microns
in diameter. They also differ in color, being reddish-brown instead of
brownish-green. The immature eggs of F. cirncscais within the con-
ceptacles are quite green, indicating that the reddish-brown pigments
develop in the eggs as they approach maturity. Later in the spring (in
April and May) gametes from the dioecious Ascophyllum nodosum were
alxi obtained in abundance. The eggs of this form are green. They
vary in diameter about between 60 and 85 microns, and occur four in a
capsule instead of eight.
Experiments were designed to see if cross-fertilizations will take
place among these three species. Fruiting tips were washed in fresh
water to kill any antherozoids which might be upon them, and were dried
with filter paper and then dipped in filtered sea water to restore the
normal salt environment. They were then placed in individual moist
chamber^ to shed.
Kggs and sperm from a given fruiting tip of /;. crcncscens often
exhibited 99 per cent fertilization. It is apparently not necessary for
the sperm of different individuals to be exchanged for the highest per-
centages of fertilization, u- in -.nine hermaphroditic animals such as the
ascidian dona (Morgan, 1924).
Cross-fertilizations were attempted as follows: /;. rcslculosus sperm
X Ascophyllum eggs, F. vcsiculosits eggs X Ascophyllum sperm, F.
vesicitlosus eggs /;. crcncsccns sperm. In all cases samples of the
eggs and sperm used were also tested against the corresponding gametes
of the same species as a control to be sun- that they were in good con-
dition. Samples of /•'. vcsiciilosus and .•isco^liylluni eggs were also run
as unfertilized controls to check against contamination. Since F. cvc-
ncsccns eggs could not be obtained separately from sperm no attempts
were made to fertilize them with foreign sperm. The F. cvcncscens
sperm could be obtained free from eggs by drawing off sea water above
evcncsccns eggs which had settled in a dish. /;. vcsiculosus sperm were
added to Ascophyllum egg> in four experiments. In the first experi-
- I ;mi indebted to Professor \V. R. Taylor for the identification of this species
in 'in a pressed specimen.
EGGS OF FUCUS AND DEVELOPMENTAL AXIS 303
n lent about ten per cent of the eggs divided, but also about ten per cent
of the unfertilized controls developed. J Cither contamination or parthe-
nogenesis occurred. The experiment was repeated three more times
with large dishes of eggs with no further development whatsoever. The
controls were good in these experiments. Ascophyllum sperm were
added to F. vesiculosus eggs in three experiments. In two cases no
divisions took place. In the third case three eggs in about 3,000 divided.
The controls were all good. F. rcsiculosns eggs were inseminated with
F. ci'ciicscens sperm in ten experiments. The controls were all good.
The percentages of development in the ten experiments were as follows :
1 per cent, 1 per cent, 0.1 per cent, 0, 0, 0, 0, 0, 0, 0.
The conclusion must be that less than 1 per cent cross-fertilization
takes place among these forms in normal sea water.
The " Group Effect "
In the course of some other experiments large numbers of dishes of
F. vesiculosus eggs had been reared in the dark and many examples were
seen of what appeared to be division of eggs so that the rhyzoids pointed
toward near neighboring spores. Since it would not be difficult to be
deceived, with neighboring cells in all directions, some sixty dishes were
prepared in which only two eggs were placed between one and three
egg-diameters apart. These were reared in the dark and were later
examined microscopically with an apparatus which made it unnecessary
to jar or touch the dishes. The results were then tabulated, dividing
the circle around each egg into quadrants, one of which included the
angle 45° to either side of the line joining the two eggs. Counts were
made .of the eggs whose rhyzoids protruded in each quadrant. The
count showed an entirely random distribution. There was no correla-
tion between the division plane and the position of the neighboring cell.
This result was very surprising as some very convincing signs of cor-
relation had been seen in dishes containing many cells. Accordingly,
dishes were prepared in which a large compact mass of eggs lay in the
center of the dish, and around about the periphery of the mass isolated
eggs were placed at intervals well apart and about two egg diameters
out from the peripheral cells of the mass. In this configuration there
is no ambiguity as to the direction of the neighboring cell mass as in a
random scattering over the bottom of a dish. The eggs were reared in
the dark. The results in these cases were as clear-cut as in those with
only two eggs in a dish, but they were quite the opposite. Some actual
counts of the directions in which the rhyzoids protruded from these out-
lying cells are as follows : Within the 180° of arc toward the central
mass 257 eggs, in the 180° away from the mass 2 eggs; in 57 eggs, out
304 D. M. WHITAKER
1 egg; in 34( >. nut 7 ; in 40, out 3 etc. The peripheral cells of the central
mass itself al-" divided with rhyzoicls inward in comparable proportions.
It appears then that some general condition resulting from large numbers
iif eggs in the dish is necessary in order that several neighbors shall have
FIG. 2. Outline sketch uf a vrmip of F. ci'cncsccns eggs reared in the dark,
• wing tlic " group effect." Some of the inner cells have divided equally without
producini: rhy/oids.
this mutual effect. The causes of this were not worked nut. For the
present ]nir])(»se it was sufficient to find a circumstance in which the
"group effect' invariably takes place. This condition is met when
eggs are placed around a central mass of hundreds or thousands of eggs
in a Syracuse dish, regardless of the shape of the mass. In very small
dishes a smaller mass mav suffice.
I-'K;. .1. Sketeh of part of an experiment, showing /•'. ci'cncsccns spores (stip-
pled ) directed liy unk-rt ilixi-d /•. Tt'siatlnsits fi'.us. These sporc^ \vi-n- reared in
the dark, and /•'. ci'cticsccns eggs are <ni <mr >idc only of tlu- /•'. i-cxicnlosns mass.
'flic eggs of .\scaph\llitin also exhibit the "group effect " and the
eggs of /•'. tTfin-sceiis sho\v it very markedly, much more markedly
than the eggs of /•". rcsiculosns. That is to say, i-ven small isolated
gmups of these egg> alone in a dish develop rhy/oids pointing toward
the nearest neighbor, or in the resultant direction if there are a number
ol neighbors. Kings of lour or five eggs ha\e the rhv/.oids all pointing
inward. In large masses not only the peripheral eggs have rhyzoicls
pointing inward, hut often five or six outer layers of cells also all point
inward. In masses of eggs of this species, especially larger masses, the
EGGS OF FUCUS AND DEVELOPMENTAL AXIS
305
innermost cells not infrequently divide into two equal and similar cells
instead of producing' one rhyzoid cell. Farmer and Williams observed
occasional equal divisions in F. vesiculosus eggs, especially when reared
in the dark. Some actual counts of the direction of the rhyzoid in
peripheral cells of masses of F. evenrscens eggs (again dividing the
circle into two divisions of 180°) are as follows: in toward center of
mass 260, out 8; in 38, out 0; in 113, out 3; in 210, out 2 etc. An
example of a small isolated cluster of these eggs is shown in Fig. 2.
To answer the two questions raised in the first part of this paper,
masses of closely-packed F. vcsiculosits eggs were arranged in Syracuse
dishes and in smaller dishes, and with a mouth pipette eggs of F. eve-
nescens were placed at intervals around the periphery of each mass.
TABLE I
The results of seven experiments showing the directive effect of unfertilized
F. vesiculosus eggs on the developing spores of F. evenescens.
Experiment No.
In
Out
Tangent
Equal Division
1
36
1
0
1
2
41
0
1
2
3
12
0
1
0
4
14
1
0
0
5
8
0
1
1
6
10
0
0
0
7
12
0
0
0
Totals
133
2
3
4
The F. evenescens eggs were placed at distances varying between one
half and two egg diameters out from the periphery of the F. vesiculosus
mass. Around the circumference the distance between consecutive F.
evenescens eggs ranged from five to a hundred or more egg diameters,
to rule out any effect which they might have on each other. (Other
tests showed that no directive influence extended beyond four or five egg
diameters.) To further control against any mutual effect of the F.
evenescens eggs, in some cases the F. evenescens eggs were placed only
on one side of the mass of vesiculosus eggs so that no other evenescens
eggs whatsoever, at any distance, would be in the direction of the central
vesiculosus mass. (Fig. 3.)
When newly shed and fertilized F. evenescens eggs were added in
place, supernumerary F. evenescens sperm also came into the dish, but
these did not fertilize the F. vesiculosus eggs of the central mass. The
dishes were then placed in a dark room under black felt and were later
examined without jarring the dishes. Since the fertilized spores adhere
21
306
D. M. WHITAKER
to the bottom df the dish they are not easily dislodged, but the unfer-
tilized Tcsicitlosus eggs are easily moved.
The results of seven experiments are given in Table I. In the
column headed " in " are represented the eggs whose rhyzoids pointed
inside an imaginary line tangent to the edge of the central mass. " Out "
represents those which pointed outside the tangent line. " Tangent " rep-
resents the eggs whose rhyzoids paralleled the tangent line. " Equal ':
represents eggs which divided equally, without producing a rhyzoid.
All four cases of equally dividing eggs developed cell plates parallel to
the tangent line, as if to protrude rhyzoids directly inward (or out).
FlG. -4. 'I In- cfitral mass consists of uniYrtili/ed n-linu /•". rcsicttlosiis eggs.
At intervals around tin- periphery are directed / . evcnescens spores, which are
stippled. Tin ter size has been slightly exaggerated. Reared in the dark.
The ^ri-at ].re]Hin«lerance of spores whirl) dixioVd \\-ith rhyzoids within
the tangent line did so with the rhyzoid pointing almost exactly toward
the nearest cell of the central mass. In experiments one, five, and six,
the /•'. < ."•< 'iicsccns eggs were placed on only one side of the central mass.
EGGS OF FUCUS AND DEVELOPMENTAL AXIS 307
In experiments two, three, four, and seven, they were placed all around
the mass. The direction of the central mass from the evenescens eggs
therefore covered all points of the compass and controlled against en-
vironmental asymmetries such as, for example, stray light (which, how-
ever, was not present). In experiments one and two the central mass
was about one centimeter across; in experiments three, four, five, six,
and seven, it was about two or three millimeters across, although the
shape of the mass was not always strictly circular. A sketch of one of
these experiments is given in Fig. 4. The peripheral evenescens cells
are placed more closely together in this experiment than in the others.
Discussion of the Directive Agency in the " Group Effect ''
The results show conclusively that eggs of F. evenescens, which tend
to divide so that the rhyzoid grows in the resultant direction of neighbor-
ing cells, are equally well directed by unfertilized resting eggs of another
species as by dividing eggs of the same species. The directive effect
therefore cannot be ascribed to any agency dependent on nuclear or cell
division in the directing cells. The effect is also non-specific, although
the two species tested are of the same genus.
I am much indebted to Professor W. J. Crozier, in whose Laboratory
these experiments were performed, for advice and criticism, and to
Professor W. H. Weston and Dr. A. E. Navez for advice and help in
locating the habitat of the Fucus and in the proper handling of the plants
in order to obtain gametes. Professor W. R. Taylor has been so kind
as to identify F. evenescens for me from a pressed specimen.
SUMMARY
1. The results of a number of observations and experiments are
presented which relate to the nature of the FUCHS egg and to some of the
changes which take place in it at fertilization.
2. Giant eggs which in some cases develop supernumerary rhyzoids
are found to originate in the fusion of single eggs within the capsule and
especially at the time of emerging from the capsule.
3. The extent to which eggs fuse is found to be greatly reduced at
low temperature and increased at high temperature.
4. Cross- fertilization between F. vesicnlosus and Ascophylluui
nodosum was found not to take place. The sperm of F. evenescens do
not fertilize the eggs of F. vesicnlosus to any appreciable extent. In-
dividual receptacles of the monoecious F. evenescens are entirely self-
fertile.
308 D. M. WHITAKER
5. The developmental axis of the spores was found to be directed
by the presence of nearby neighbors in F. rcsicitlosus, F. cvencsccns, and
in Ascuphyllum. The first division plane tends to lie so that the rhyzoid
pnitriuk's in the resultant direction of near neighbors. No jelly or other
solid egg substance traverses the space between affected eggs.
6. Unfertilized resting eggs of another species were found to direct
the division planes of eggs of F. crcncsccns equally well as dividing eggs
of the same species. Therefore the directive effect cannot be ascribed to
any agency dependent on nuclear or cell division in the directing cells.
BIBLIOGRAPHY
FARMKK, T. E., AND J. L. WILLIAMS, 1896. On Fertilization, and the Segmenta-
tion of the Spore, in Fucus. Proc. Roy. Soc., 60: 188.
FAKMKR, J. B., AND J. L. WILLIAMS, 1898. Contributions to Our Knowledge of
the Fucacere : Their Life-history and Cytology. Phil. Trans. Roy. Soc.
B., 190: 623.
Hoi LAI XIIKR, A., AND E. ScHOEFFEL, 1931. Mitogcnctic Rays. Quart. Rev. Biol.,
6: 215.
HTRD, AXXIE MAV, 1920. Effect of Unilateral Monochromatic Light and Group
Orientation on the Polarity of Germinating Fucus Spores. Dot. Gaz., 70:
25.
I.rxn, E. J., 1923. Electrical Control of Organic Polarity in the Egg of Fucus.
Dot. Gaz., 76: 288.
MIIRCAN, T. H., 1924. Self-Fertility in Ciona in Relation to Cross-fertility. Jour.
Exper. Zoo/., 40: 301.
Mok(,.\x, T. H., 1928. Experimental Embryology. Columbia University Press.
I'IKRCE, G. J., 1906. Studies of Irritability in Plants. Ann. Bof.. 20: 449.
\VIXKLER, H., 1900. Ueber den Einfluss ausserer Factoren auf die Theilung der
Eier von Cystosira barbata. Bcr. Dcntsch. Dot. Gesells., 18: 297.
YAMANOITHF, SHH.KO, 1909. Mitosis in Fucus. Bot. Gaz., 47: 173.
SPERMATOGENESIS IN THE CALIFORNIA OYSTER
(OSTREA LURID A)
WESLEY R. COE
OSBORN ZOOLOGICAL LABORATORY, YALE UNIVERSITY
In connection with an investigation on the sequence of sexual phases
in this species (Coe, 1931) numerous preparations were made of the
gonads of oysters of definitely known ages. Some of these illustrate
very clearly the general features of the processes concerned in the forma-
tion of the gametes. And since no very precise description of these
processes has been published for any of the numerous species of oysters,
it is hoped that this brief paper may be helpful in bringing to light some
interesting deviations from the more usual types of spermatogenesis.
Nearly a half century ago Hoek (1883) published an excellent gen-
eral account of the origin and growth of the gonads in Ostrea cdulis,
but he was unable to follow the cellular changes which occvir in spermat-
ogenesis. In the several papers by Orton, especially in the more recent
extensive studies (1926) on the sex change in O. cdulis, is much infor-
mation on the characteristics of the gonads in each of the sexual phases.
But these studies have not included gametogenesis.
It is well known that the spermatozoa of hermaphroditic species of
oysters leave the body in the form of balls or ellipsoid clusters of closely
packed ripe sperm cells. Each ball usually consists of from 250 to 2000
or more spermatozoa, each with its head directed toward the center of
the ball and with its flagellum extending radially above the surface
(Figs. 1, 4).
These balls are formed in the gonads and pass through the genital
ducts into the mantle cavity, and thence out of the body, in great num-
bers whenever the animal in the ripe male phase is suitably stimulated.
In this connection it should be stated that all individuals of this
species of oyster are protandric, and that there is a rhythmical alternation
of female and male phases throughout the remainder of their lives (Coe,
1931). There are many intergrading stages in the change from one
sex phase to the other, particularly in young animals. Furthermore, it
frequently happens that one part of the reproductive system reaches a
certain phase of sexuality in advance of other parts, whereby one por-
tion of the system will have the characteristics of one sex while the rest
of the gonads are predominantly of the other sex.
309
310
W. R. COE
All grades of hermaphroditism are thus found in an oyster popula-
tion at all seasons of the year. As a rule ahout half of the population
niav In- roughlv classed as intersexual forms, an equal number heing pre-
dominantly of one sexual phase or the other. But animals exclusively
male or female are few in number at any season, particularly those that
are exclusively female.
The vast majority of those in the female phase either have more or
less abundant sperm balls, remaining from the preceding male phase, in
the follicles of the gonads or in the genital canals, or else they show some
follicles in which the spermatogenesis for the succeeding male phase is
anticipated. The ripe ova often mingle freely with the sperm balls in
mpt
•
...-,-
• w # • "
ov .85. \
FlG. I. Diagrams of portions of ripe gonads in first and third male phases
drawn to the same ^calr. ./, primary gonad, climax of first male phase, with ripe
sperm-balls (.?/>") filling both lumen and ciliated genital canal (</<•); follicle bor-
dered with closely placed ovocytes {ov} anticipatory of fir>t female phase. B, por-
tion of secondary gonad, indicating it* much gn iter size at climax of third male
phase; very numerous balls (sf>z) now fill the much larger follicle and genital
canal (gc) ; spermatocytes (spc) and a few large ovocytes (ov) border the lumen.
the genital ducts, but the firm attachment of each spermatozoon prevents
self-fertilization, at least until after the sperm have been discharged into
the mantle cavity.
In the male phase similarly, particularly in young males, the follicles
of the ^onads usually contain more or less numerous ovo^onia or well-
grown ovocytes or both (Fig. I). Only in the oldest animals are the
transition stages almost eliminated.
In thf young animal the first traces of the gonads appear at the age
of about eight weeks. The few cells composing these gonads show no
distinguishing characteristics of sexual differentiation, but at the age
of twelve to sixteen werUs each gonad in exerv animal studied shows that
SPERMATOGENESIS IN OSTkl-A I.URIDA
311
both primitive ovogonia and spermatogonia are present. The spermato-
gonia, however, proliferate more rapidly than do the ovogonia and the
gonad soon acquires the characteristics of a spermary although ovogonia
and ovocytes are always present. Spermatogenesis quickly follows if
the temperature is sufficiently warm and the ripe spermatozoa are ready
to be discharged when the oyster is about five months of age (Fig. 2).
Before this initial male phase has been completed and before any of
the sperm-balls have been discharged, the proliferation of the ovogonia
and their transformation into ovocytes are in progress. Most of the
FIG. 2. Diagram of successive stages in spermatogenesis. A, two indifferent
germ cells on wall of gonad ; B, small group of spermatogonia, with reticular chro-
matin and conspicuous nucleoli ; C, small group of secondary spermatogonia ; D, di-
vision of secondary spermatogonia to form spermatocytes ; E, primary spermato-
cytes with slender chromosomes ; F-K, division of primary spermatocytes ; L-R,
division of secondary spermatocyte ; S, T, transformation of spermatid into the
mature spermatozoon.
sperm-balls are then discharged from the body, whereupon the animal
assumes the first female phase, although some sperm-balls are always
left in the genital ducts, and many spermatogonia for the subsequent
male phase are present in the gonads.
In the female phase the ovocytes build up their yolk materials and
ovulation occurs at the age of about six months. The eggs are retained
in the mantle cavity of the parent during fertilization and cleavage and
through development until the embryos have become provided with a
bivalved, straight-hinged shell.
While the embryos are developing within the mantle cavity the
spermatogonia remaining in the gonads begin a rapid spermatogenesis
and even by the time the embryos have been spawned the second male
312 W. R. COE
phase has been reached. The number of sperm-balls produced is now
vastly greater than in the first male phase and a much greater proportion
of them contain the maximum number of spermatozoa. If the animal is
well nourished some hundreds of thousands of such sperm masses are
formed, with upwards of 2000 spermatozoa in each.
After the ripening and discharge of the sperm will come a recupera-
tion period. And, apparently, these alternating sexual phases will be
repeated regularly throughout the remainder of the animal's life. But
it is not at all improbable that in certain individuals, and possibly in some
hereditary strains, one sexual phase or the other may be considerably
reduced in older animals, with a corresponding tendency toward a di-
oecious condition. Furthermore, if the nutritional conditions are favor-
able the recuperation period may be abbreviated or eliminated, resulting
in several changes of the sexual phase in a single breeding season. Or
a recuperation interval may divide any of the male phases, after the
first, into two separate parts, one period of spermatogenesis immediately
following ovulation and the other preceding the next female phase.
SPERMATOGENESIS
The successive stages in spermatogenesis will be discussed in the
order in which they appear in the gonads of the young animal.
Indifferent gonia. — The earliest gonads, as found in young animals
about eight weeks after attachment, consist of only a small number of
cells and these show no recognizable characteristics that might indicate
to which sex line they are ultimately destined. A few weeks later, how-
ever, after a large number of descendants has been produced, the two
types of gonia are easily recognizable as such. The ovogonia then lie in
a single row close beneath the surface of the gonad. while the primary
spermatogonia occur singly or in small groups either against the wall or
separated from it by several ovogonia (Figs. 1, 2).
Primary spermatogonia. — Eaeh of the balls or clusters of ripe sper-
matozoa is derived from a .single primary spermatogonium, and as there
are commonly from _'50 to 2000 or more spermatozoa in each ball the
primary spermatogonium must divide six to nine times to produce the
M to 500 or more secondary spermatogonia re<|iiired. The number of
divisions presumably depends upon the amount of nourishment available.
The first division of the primary spermatogonium is frequently verti-
cal to the surface of the gonad, one of the two daughter cells remaining
in contact with the surface (Fig. 2). Successive divisions result first in
a morula of cells and then in a more or less regular spherical mass.
i'.ach of the constituent cells then assumes a p\ ramidal shape, with the
apex toward the center of the group and with the nucleus near the base
SPERMATOGENESIS IN OSTREA LURIDA
313
of the cell, that is, near the surface of the sphere (Figs. 2, 4). With
still greater multiplication in numbers some of these secondary spermato-
gonia become crowded into the center of the sphere, causing a change in
the shape of such as remain in contact with the surface. All the sper-
matogonia have large, vesicular nuclei, each with a conspicuous nucleolus
and loose chromatin reticulum. Although the cell bodies are in close
contact and are held together by a delicate non-cellular secretion, the
cytoplasm of adjacent cells is always more or less completely separated
(Fig. 2). In less well preserved specimens, however, the clusters may
have the appearance of syncytial masses with nuclei imbedded radially
in the common protoplasmic matrix.
F G H I J
FIG. 3. Transformation and division of primary spermatocyte. A, leptotene;
B, C, D, prophase groups of chromosomes ; E-J ' , mitosis and formation of sec-
ondary spermatocytes.
Except at the time of mitosis the spermatogonia must absorb a con-
siderable amount of nourishment, for the final cells remaining after the
spermatogonial divisions have been completed are about one-eighth as
large as was the original spermatogpnium. The latter measures about
.0057 mm. in diameter in the prepared sections while the final spermato-
gonia are about .0028 mm. in diameter. Intermediate gonia are inter-
mediate in size.
Primary spermatocytes. — Following the last spermatogonial division
the resulting primary spermatocytes are retained in the same spherical
groups. Very little growth takes place, the nuclei soon showing the
chromosomes in slender spiremes, apparently followed by the usual pro-
cess of synapsis (Fig. 2). The typical brachytene stage soon appears
and then the prochromosomes are arranged close beneath the nuclear
membrane (Fig. 3). Prophase, metaphase, telophase are all of typicaJ
314
W. R. COE
appeanmcr. with a delicate spindle of the usual form. The chromosome
numlKT could not be definitely determined because of the crowded con-
dition of tlu- metapha.se and anaphase plates, hut it is not very large.
There arc two typical spermatocyte divisions.
Si-coinlury spcrniatocytes. — The secondary spermatocytcs arc like-
wise hi'ld together in a crowded, spherical mass. Nuclear behavior and
mitotic figures do not deviate from the typical condition (Figs. 2, 4).
Spermatids. — These also remain in close contact and become defi-
nitelv oriented, each with its longer axis in a radial position in the
irregularly spherical mass of from 250 to 2000 or more similar cells
which compose the sperm-ball (Fig. 4).
Fi<;. 4. Sta.m^ in formation of spcrm-l)all. A, group of young spermatogonia ;
n. later stage of same group; C, primary spermatocytes; D, division into secondary
spermatocyh ^ ; /:, *pennatiil> ; / , ripr -jn i in-hall, with radiating flagclla of the
spermatozoa.
Spermatozoa. — During the trans formation of the spermatid to the
-peniirito/ooii the llagellum gr(j\\'S out radially and proji-cts far beyond
the surface of the group. The free outward growth of the flagellum
shows that the sperm-ball has only a tenuous or gelatinous covering and
that each spermalo/.non is held in place by a common matrix of trans-
lucent gelatinous secretion.
K\c.],t for -uch relatively few groups of spermatogonia as have
retained positions in contact with the wall of the gonad, the development
of the sperm-balls has taken place in the fluid contained within the
lumen. I'sually the hist sperm-balls to be completed arc those adjacent
SPERMATOGENESIS IN OSTREA LURIDA 315
to the spacious portion of the gonad which is continuous with the cili-
ated genital canals. Plere they and the later ones accumulate until many
thousands and in some cases perhaps hundreds of thousands are ready
to he discharged (Fig. 1). The sixe of the halls varies greatly, due to
variation in the nuinher of the constituent spermatozoa, but most of
them are between .04 and .06 mm. in longest diameter.
The male phase has now reached its climax and upon a suitable stim-
ulus, such as a rise of temperature from below the critical point of 16°
C. to above that point in the spring or, presumably, by the presence of
eggs of other individuals in the vicinity at other times during the seven
months of the breeding season, the spasmodic contractions of the oyster's
body forces the myriads of sperm-balls into the water.
On reaching the sea water the sperm-balls rotate rapidly, due to the
lashing of the tails of all the contained spermatozoa. The cementing
substance is gradually dissolved, liberating the spermatozoa which are
then free to swim about in the water. The ripe spermatozoon is not
much more than .0012 mm. in diameter, with a flagellum about twenty
times as long as the rounded oval head (Fig. 2). After all the sper-
matozoa have worked themselves free, the remainder of the intercellular
matrix of the sperm-ball is left behind as an amorphous gelatinous
material.
Comparison of the gonads of 0. virginica with those of 0. lurida
shows that in both species there is a close agreement in the general fea-
tures of spermatogenesis. But in the former species the derivatives
both of the primary spermatogonia and of the spermatocytes separate
freely, so that there is no aggregation into masses other than the asso-
ciation which results from the proximity of neighboring cells. The
young spermatozoa are thus free to move individually in the lumens of
the gonads and in the genital ducts, in marked contrast with those of
O. lurida, where a special adaptation prevents, or diminishes the oppor-
tunities for, self-fertilization.
LITERATURE
COE, W. R., 1931. Sexua! Rhythm in the California Oyster (O. lurida). Science,
74: 247-249.
HOEK, P. P. C., 1833-84. De voortplantingsorganen van de oester : les organes cle
la generation de 1'huitre. Tijdrsclir. Ncdcrl. Dicrkitndiye Vcr., 1: Suppl.,
113-253.
ORTON, J. H., 1926-27. Observations and Experiments on Sex Change in the
European Oyster (O. edulis). Jour. Mar. Biol. Assn., 14: 967-1045.
STUDIES OF PHOTODYNAMIC ACTION
III. THE DIFFERENCE ix MECHANISM BETWEEN PHOTODYNAMIC
HEMOLYSIS AND HF.MOLYSIS BY NON-IRRADIATED EOSINE
H. F. BLUM AND G. C. McBRIDE
(From the Division of Physiology, University of California Medical School,
Berkeley, California)
Dyes which bring about photodynamic hemolysis, in many instances,
bring about the same effect in the absence of light (Dunkehvirkung),
when the dye is present in sufficiently high concentration. This sug-
gests the possibility that a reaction of the dye with cell constituents
which is independent of light underlies the hemolysis; and that this
reaction is accelerated by the activation of dye molecules by absorbed
radiation, with the result that Ivmolysis occurs in lower concentrations
of the dye. Certain rough correlations between the effect of the non-
irradiated dyes and the photodynamic effects have been pointed out by
Jodlbauer and Haffner (1921a) and by Blum (1930/?) which would
support this thesis, but it is possible that hemolysis is initiated by en-
tirely different mechanisms in the two cases. Photodynamic hemolysis
has been shown to require tin- presence of molecular oxygen (Hassel-
balch. 1909; Schmidt and Xorman, 1922). and there seems little doubt
that this phenomenon is dependent upon oxidations by molecular oxygen
activated in some way by light. Obviously, if the hemolysis produced
by the non-irradiated dye is dependent upon the same reactions, it must
likewise be inhibited by the absence of molecular oxygen. The attempt
to separate the two processes on this basis has been the object of the
following experiments.
EXPERIMENTAL
Quantitative experimental treatment of ibis problem meets with
various difficulties. Measurement and comparison of the oxygen con-
sumption during hemolysis by irradiated and non-irradiated dyes meets
the </ priori objection that certain dyes greatly alter the normal metabol-
ism of cells without apparent destructive' effects (see I'arron and Hoff-
man. 1'MO), which might result in faKe conclusions as to the oxygen
consumption of the process leading to hemolvsis. Reducing the partial
pressure of oxygen in any way, with the object of studying the effect on
the hemolytic process, brings about changes in hydrogen ion concentra-
316
PHOTODYNAMIC ACTION STUDIES 317
tion within the red blood cell due to the formation of reduced hemo-
globin. This change in hydrogen ion concentration may considerably
affect the hemolytic process (Jodlbauer and Haffner, 1921ft; Blum,
1930ft), without reference to oxidative reactions. Such factors offer
considerable difficulties in quantitative experimentation, and it has,
therefore, appeared wise to attempt only to demonstrate qualitatively,
whether or not hemolysis by dyes may occur in the absence of light and
oxygen.
The method employed has been as follows : Series of cosine solutions
were prepared covering a range of concentrations which included the
minimum concentration found to bring about hemolysis in air in the
dark. The solutions were made up with isosmotic phosphate buffers
usually at pH 7.0, according to the procedure described by Blum
(1930a). Suspensions of 0.5 per cent red blood cells were made with
these solutions, oxygen removed, and one series exposed to sunlight,
while the other was maintained in the dark. If hemolysis by the non-ir-
radiated dye, as wrell as photodynamic hemolysis, requires oxygen, he-
molysis should not appear at any dye concentration in either the irradiated
or the non-irradiated series. On the other hand, if the action of the non-
irradiated dye does not require oxygen, there should be a concentration
in each series above which hemolysis should occur. In the latter case
the minimum concentration at which hemolysis occurs need not be ex-
actly the same as the corresponding minimum in air, since the removal
of oxygen would result in a change in the hydrogen ion concentration
within the cell which might cause a shift in this minimum.
The validity of the results obtained in this way depends upon the
removal of oxygen to a level which will not allow the oxidative mecha-
nisms leading to hemolysis to proceed at a demonstrable rate. The only
criterion for this is the complete inhibition of hemolysis in the irradiated
systems. This was found difficult, the difficulty lying apparently in the
removal of the oxygen from the cells themselves. The suspensions
which we have found convenient for the study of the hemolytic process
contain 0.5 per cent red blood cells, and it may be readily calculated that
in such a system the cells contain about one- fourth of the total oxygen
in the system. The cells contain about forty per cent by volume of
oxygen, or 0.2 cc. per 100 cc. of suspension containing 0.5 per cent cells.
From the absorption coefficient of oxygen in water approximately three
cc. of oxygen are absorbed in 100 cc. of water at 25° C., and since our
solutions are saturated with air and not with oxygen, they should^ con-
tain approximately one-fifth of this quantity or 0.6 cc. per 100 cubic
centimeters. Thus there is, roughly, one-third as much loosely-bound
oxygen in the cells as there is dissolved oxygen in the surrounding solu-
318
II. F. r.I.UM AXD G. C. McBRIDE
tion. It is thus apparent that the oxygen must be removed from the
cell as well as fnun the solution in order to establish the desired low
gen tm>ion. It was not found possible to completely inhibit the
effects of the irradiated dye by reducing the atmospheric pressure above
tlu- solution. It was likewise found difficult to obtain conclusive results
bv attempting to remove the oxygen by bubbling nitrogen through the
solutions for a considerable time. It was found, however, that definite
-ults could IK- obtained by using carbon monoxide to remove the oxy-
gen from the cells. Tin- procedure in these cases was first to bubble
nitrogen through a series of tubes, each containing 2 cc. of dye solution
of a given concentration without cells for 15 to 20 minutes to ensure
the removal of oxygen from the solutions. Carbon monoxide was
bubbled through a 50 per cent suspension of washed red blood cells to
TABLE I
Irradiated systems exposed to mid-day sunlight for 1 hour (12:15 p.m.-l :15 p.m.
August 21, 1931). Observations made at the end of 6 hours following mixing of cells
with dye solution. H == compleie hemulysi^; dJ] pariial hrnmlvMs. Solutions
contain sodium phosphate buffer isosmotic with 0.15 M Nad, pi I 7.0, + 0.5 per cent
r. b. c. Human.
c. Systems in
a. Systems in Air
I). Systems in ( ( )
CO 80
1 . -in
02 = 20
t ration
Irra-
\>.i Irra-
Irra-
Not Irru-
Irra-
\<>t Irra-
diated
di.r
diated
diated
diated
••nt
1 t
II
II
H
(H)
II
II
.7
II
H
II
II
II
(H)
.35
H
—
—
—
(H)
—
.175
(H)
—
—
—
H
—
.Os 7
(II)
—
—
—
(H)
—
.011
II
—
—
—
II
—
remove the- oxygen from these; 0.2 cc. of this suspension was then added
to each tube to form a 0.5 per cent suspension of cells, the tubes being
opened to tin- air for as short a time as possible in order to avoid the
entrance of oxx^eu. The suspensions were then Hushed out with about
400 cc. of carbon monoxide. Such treatment was found very effective
in inhibiting the photo-reaction, but. as will be seen by reference to
Table I, did not inhibit the action of the non-irradiated dye. In order
to rule out any possible specific effect of carbon monoxide, similar sys-
tems were treated with a mixture of 20 per cent oxygen and XO per cent
carbon monoxide.
Table I presents the results of a typical experiment, in which three
series of cell suspensions were exposed to sunlight, (a) in air, (b) in
PHOTODYNAMIC ACTION STUDIES 319
carbon monoxide, and (c) in a mixture of carbon monoxide and oxy-
gen, while three similar series \vere maintained in the dark. The re-
sults show that the photodynamic effects are completely inhibited by the
absence of oxygen, while the effects of the non-irradiated dye are not.
The fact that the photo-effect is completely inhibited in an atmosphere
of carbon monoxide indicates that the oxygen content of the system has
been lowered sufficiently so that hemolysis by the non-irradiated dye
should also be inhibited if it is dependent upon the same oxidative re-
actions as the photodynamic effect. The fact that carbon monoxide
does not inhibit the light reaction when oxygen is present indicates that
the inhibition is not a specific action of the carbon monoxide but is due
to lack of oxygen. While the results of such experiments vary some-
what with regard to the rate of development of hemolysis, in no case
has it been possible to completely inhibit hemolysis by non-irradiated
dyes. A certain amount of the variation may be due to temperature
differences. The non-irradiated systems were maintained at a tempera-
ture of approximately 25° C. during the period before the observations
were made. On the other hand, the irradiated systems were exposed
during the one-hour period of the irradiation to a variable temperature,
which, howrever, was never higher than 27° C. and in some experiments
was considerably lower than the temperature of the non-irradiated sys-
tems. So far as can be determined, by such qualitative observation as
we have used, the rate of hemolysis is somewhat decreased in the sys-
tems in contact with CO or mixtures of CO and On. It seems probable
that this is due to the difference in hydrogen ion concentration of the
cells containing carboxyhemoglobin from that of those containing oxy-
hemoglobin. It is possible, of course, that in the series exposed to
light, the carboxyhemoglobin is dissociated to some extent by the action
of light with the formation of oxyhemoglobin or reduced hemoglobin,
depending upon whether oxygen is present or not. This might account
for some differences in the rate of hemolysis between irradiated and
non-irradiated systems.
The fact that hemolysis is completely inhibited in the absence of
oxygen excludes the possibility that reactions of the type described by
Levaillant (1923) and Windaus and Borgeaud (1928) may bring about
the destructive changes leading to hemolysis. These reactions take place
in the absence of oxygen and may be considered as oxidations in which
the dye acts as a hydrogen acceptor. The dye is reduced in these cases
to the colorless leucobase, and the fact that no bleaching of the dye was
observed in our experiments indicates that reactions of this type did
not occur to any appreciable extent.
320 H. F. BLUM AND G. C. McBRIDE
\Vhile it was not found possible, as mentioned above, to inhibit the
photo-effect completely by evacuation or by bubbling nitrogen through
the suspensions, it was found that such treatment markedly decreased
the photo-effect, but did not alter the hemolytic effect of the non-
irradiated dye. Similar results were observed when the attempt was
made to inhibit these reactions by the use of reducing agents in the
solution. Experiments were carried out using Na2SO3, Na,SoOn, and
Xa\'( ).. in concentrations up to 0.1M. It was found impossible to com-
pletely inhibit the photo-reaction by means of these reducing agents, and
therefore it cannot be assumed that the reducing power of the solution
is sufficient to completely inhibit the action of the non-irradiated dye if
it involves the same process as the photo-reaction. However, the fact
that these reagents have no apparent tendency to inhibit the dark re-
action indicates that the two processes are essentially different. Thus,
while the above evidence may not, perhaps, be considered as absolutely
conclusive, collectively it gives strong support to the view that the dark
reaction is not an oxidation by molecular oxygen, whereas the photo-
reaction is.
DISCUSSION
Hasselbalch (1909) performed experiments to test whether the
hemolysis by non-irradiated dyes requires oxygen, and reported results
contrary to those described above. Using red blood cells in suspension
in isotonic XaCl solution which contained a given concentration of dye
-ufticicnt to bring about hemolysis in the absence of light, he found that
the evacuation of the air above the solutions inhibited the hemolysis.
Evacuation under these conditions would remove CO., as well as O2,
and since the solutions in which the cells were suspended were un-
buffered, the removal of CO., must have decreased their hydrogen ion
concentration. Furthermore, the removal of both CO., and O, from
the cells themselves must have resulted in a decrease of hydrogen ion
concentration within the cells due to formation of reduced hemoglobin,
llemolvsis bv non-irradiated fluorescein dves is markedly affected by
•• f f j
hydrogen ion concentration (Jodlbauer and Ilaffner, \92lb; Blum,
1930a), the minimum concentration necessary to bring about hemolysis
increasing as the hydrogen ion concentration decreases. Thus it seems
quite probable that the results obtained by I lasselbalch with fluorescein
• lyes (cosine and rose bengal) were due to the decrease of hydrogen ion
concentration to a value at which the concentration of the dye employed
would not produce hemolysis in the absence of light. The justification
of this criticism will appear upon the examination of the tables given
PHOTODYNAMIC ACTION STUDIES 321
by Blum (1930/?).1 Hasselbalch also used quinine hydrochloricle and
quinine bisulphate, finding that evacuation prevented hcmolysis in the
former but not in the latter case. This variation in effect indicates that
the factor affecting hemolysis was probably something other than the
CX content of the system. Although, as stated above, changes of hydro-
gen ion concentration must have occurred in our systems due to the
formation of carboxyhemoglobin, this did not mask the effect of O2
lack because the observations were made over a wide range of dye
concentrations.
EXPERIMENTAL — -THE EFFECT OF CYANIDE ON PHOTODYNAMIC
HEMOLYSIS
It has been suggested that the mechanism of photodynamic action
involves the normal respiratory mechanisms of the cells themselves
(e.g., Metzner, 1919 and 1921). If this were true, photodynamic he-
molysis should be inhibited by the inhibition of the respiratory enzymes.
In the above experiments it was found that the photodynamic effects
are not inhibited by CO provided O, is present. The presence of CO
should partially inhibit all the respiratory oxidative mechanisms of the
cell except the aerobic dehydrases.2 However, light decreases the in-
hibitory effect of CO on certain of these mechanisms (" respiratory en-
zyme " of Warburg, " indophenol oxidase " of Keilin), and it is pos-
sible that the inhibitory effect of CO was very slight in the systems
where mixtures of CO and O, were used (Warburg, 1926).
To test this question further, cyanide was used to inhibit respiratory
enzymes. Series of dilutions of cosine were prepared as above, to a
part of which M/100 KCN was added. Red blood cells were added
(0.5 per cent) and a part of the KCN series was exposed to light to-
gether with control series not containing KCN ; other KCN and control
series were maintained in the dark. In no case could a difference be
detected between the KCN series and the controls in either the irradi-
ated or the non-irradiated systems. These experiments are in agree-
ment with those of Loeb (1907) and Moore (1928), who found that
KCN did not inhibit destructive changes in echinoderm eggs by cosine
and sunlight, and of Baumberger et al. (1929), who found that cyanide
did not inhibit the photodynamic action of methylene blue in preventing
the clotting of blood plasma. Cooke and Loeb (1909) found that KCN
1 The values for molar concentrations of dye given in these tables are in error;
the decimal point should in all cases be moved one place to the right. Hasselbalch
used At/200 cosine in his experiments, and it will be seen that this concentration is,
according to these tables, one at which a small difference in hydrogen ion con-
centration might determine the occurrence or non-occurrence of hemolysis.
-The nomenclature here used is that of Dixon (1929).
22
II. F. BLUM AXD G. C. McBRIDE
increased the photodynamic effects of some dyes on eggs, but this may
have been due to hydrogen ion concentration effects.
The addition of M/100 KCN should serve to inhibit markedly all
the known respiratory mechanisms with the exception of the aerobic
dehydrases, including those in which light interferes with inhibition by
carbon monoxide. If these mechanisms played a part in the production
of photodynamic hemolysis, the inhibitory effect of the cyanide should
be reflected in a reduction of hemolysis. As stated above, no such de-
crease could be observed.
While it is possible that the aerobic dehydrases may play a part, it
seems probable that the photodynamic effects are the result of direct
oxidation of cell constituents by molecular oxygen, the activation of the
OL, resulting from light energy absorbed by a sensitizer and completely
independent of activation by cellular enzymes. The destruction of
respiratory cn/ymcs might play a more important part in cells in which
respiration is more active than in red blood cells, and may possibly ac-
count for the induced tropisms of Metzner (1919, 1921), as he suggests,
but this explanation has no experimental support.
These experiments also suggest that hydrogen peroxide is not
formed as an intermediate step in photodynamic action. If H..O, took
a part in the oxidations, catalase should tend to oppose the photodynamic
effect by its destruction ; in such a case the inhibition of catalase by cy-
anide should result in increased photodynamic effects. As stated above,
cyanide has no effect whatsoever on photodynamic hemolysis, and since
this is true, it appears improbable that I-LCX is formed as an intermedi-
ate. This does not, howexer. deny the formation of intermediate or-
ganic peroxides which would not be attacked by catalase.
Sr.M MARY
1. The absence of molecular oxygen completely inhibits photo-
dynamic hemolysis but does not inhibit the hcmolytic action of the
non-irradiated dye. The two phenomena are thus dependent upon
different fundamental mechanisms.
2. Cvanide does not inhibit hemolysis either by the irradiated or
non-irradiated dye. Thus the respiratory mechanisms of the cell, with
the exception of the aerobic dehydrases, cannot play a part in the pro-
duction of photodynamic hemolysis.
PHOTODYNAMIC ACTION STUDIES
BIBLIOGRAPHY
BARRON, E. S. G., AND L. A. HOFFMAN, 1930. The Catalytic Effect of Dyes on
the Oxygen Consumption of Living Cells. Jour. Gen. Physio!., 13: 483.
BAUMBERGER, J. P., R. T. BIGOTTI, AND K. BARDWELL, 1929. The Photodynamic
Action of Methylene Blue on the Clotting Process. Proc. XHIth Int.
Physiol. Cong., Am. Jour. Physiol., 90: 277.
BLUM, H. F., 1930a. Studies of Photodynamic Action. I. Hemolysis by previ-
ously irradiated fluorescein dyes. Biol. Bull., 58: 224.
BLUM, H. F., 1930/>. Studies of Photodynamic Action. II. The relationship be-
tween hemolysis by irradiated and non-irradiated cosine. Biol. Bull., 59:
81.
COOKE, E., AND L. LOF.B, 1909. Uber die Giftigkeit einiger Farbstoffe fur die Eier
von Asterias und von Fundulus. Biochcm. Zcitschr., 20: 167.
DIXON, M., 1929. Oxidation Mechanisms in Animal Tissues. Biol. Rev., 4: 352.
HASSELBALCH, K. A., 1909. Untersuchungen iiber die Wirkung des Lichtes auf
Blutfarbstoffe und rote Blutkorperchen wie auch iiber optische Sensibilisa-
tion fiir diese Lichtwirkungen. Biochcm. Zcitschr., 19: 435.
JODLBAUER, A., AND F. HAFFNER, 1921a. Uber den Zussamenhang von Dunkel-
wirkung fluorescierender Stoffe und Photodynamie auf Zellen. Biochem.
Zeitschr., 118: 150.
JODLBAUER, A., AND F. HAFFNF.R, 192 Ib. liber die Wirkung von Eosin und Rose
bengale auf rote Blutkorperchen und den Zussamenhang von Aufnahme
und biologischer Wirkung. Pfliigcr's Arch., 189: 243.
LEVAILLANT, R., 1923. Fluorescence et Photochimie. Compt. Rend. Acad. Sci.,
177: 398.
LOEB, L., 1907. Uber den Einfluss des Lichtes auf die Farbung und die Ent-
wicklung von Eiern von Asterias in Losungen verchiedener Farbstoffe.
Arch. f. Entivick.-mech., 23: 359.
METZNER, P., 1919. Uber die Wirkung photodynamischer Stoffe auf Spirillum
volutans und die Beziehungen der photodynamischen Erscheinung zur
Phototaxis. Biochcm. Zeitschr., 101: 33.
METZNER, P., 1921. Zur Kenntnis der photodynamischen Erscheinung : Die in-
duzierte Phototaxis bei Paramecium caudatum. Biochcm. Zcitschr., 113:
145.
MOORE, A. R., 1928. Photodynamic Effects of Eosin on the Eggs of the Sea
Urchin, Strongylocentrotus purpuratus. Arch, di Sci. Biol., 12: 231.
SCHMIDT, C. L. A., AND G. F. NORMAN, 1922. Further Studies on Eosin Hemoly-
sis. Jour. Gen. Physiol., 4: 681.
WARBURG, O., 1926. Uber die Wirkung des Kohlenoxyds auf den Stoffwechsel
der Hefe. Biochcm. Zeitschr., 177: 471.
WINDAUS, A., AND P. BoRGEAUD, 1928. Uber die photochemische Dehydrierung
des Ergosterins. Licbig's Ann., 460: 235.
THK MATURATION DIVISIONS AND SEGREGATION OF
HETEROMORPHIC HOMOLOGOUS CHROMOSOMES
IN ACRIDIDAE (ORTHOPTERA)
E. ELEANOR CAROTHKRS
DEPARTMENT OF ZOOLOGY, UNIVERSITY OF PENNSYLVANIA
CONTEXTS
I. Introduction 324
1 . Definition of a chromosome 325
2. Time of segregation 325
,x The historical!}' correct usage of the term maturation as applied
1" uanirt' 'Lienesis 329
II. Material and acknowledgments 331
III. Observations 332
1. Segregation of unequal homologues 332
( 1 ) Triincrotropis citrina 332
(2) Mecostethus (jracilis 338
( .> ) Amphitornus tricolor 338
2. Chromomere vesicles and the origin of unequal homologues 331)
.v An accident during mitosis, .•/. tricolor 341
4. Incipient cictad formation, T. citrina 341
IV. I >i>i -n^sion 347
\~. Literature list 348
VI. Explanation of plates
VII. PI.
I. IXTROI>V< TK>\
The presentation of scientific facts to students and to the public
requires coordinated efforts between research workers on the one hand
and authors of text-books and popular articles on the other. Cer-
tainly, the duty of an investigator, when a group of facts sufficient to
justify a conclusion has been definitely a-crrtained, is to present both
the tin ts .UK! the conclusion in a clear and concise form through the
proper channels. But the obligation of one who prepares a text-book
i- no less to keep diligently in touch with these sources of information
.ind to present adequately and accurately the various subjects with
which he deals.
The whole field of biology presents no more clear, simple and
beautifully logical process than that of maturation. Vet, judging from
the pre-entat ion of this important subject in current text-books and the
conceptions derived from these presentations by university students,
324
MATURATION AND SEGREGATION 325
someone has failed in his duty. Serious consideration of the problem
of responsibility leads me to the conclusion that the blame lies largely
with the cytologists. The facts necessary to a clear comprehension of
the mechanism of maturation have been available for a number of
years though certainly not in a particularly clear or readily accessible
form, when one considers the amount of attention the author of a gen-
eral text-book can give to the subject.
The difficulties seem to be due chiefly to hazy conceptions concern-
ing three points: (1) Definition of a chromosome; (2) time of segregation
and (3) the historically correct usage of the term maturation as applied
to gametogenesis. These will be considered in turn.
1 . Definition of a Chromosome. — Little excuse exists for the misuse of
this word. Waldeyer (1888) named and defined chromosomes as the
individual, rod or loop-shaped, longitudinally split, basophilic bodies
which are formed from the nuclear network during mitosis. McClung
(1905) amplified this definition as follows: "Chromosomes are chro-
matic elements acting as unit structures during mitosis. Chromo-
somes are of two general classes:
1. Simple — containing two chromatids in metaphase.
2. Multiple — containing more than two chromatids in metaphase and
formed by the union of simple chromosomes.
(a) Tetrads, containing four chromatids (derived from a pair of
homologues, as ordinarily used.)
(b) Hexads, containing six chromatids.
(c*) Octads, containing eight chromatids (etc.)."
"A chromatid is a (longitudinal) half of a simple chromosome."
The parts in parenthesis are mine. In other words, chromosomes are
the individual, chromatic elements which appear definitely in the
nucleus at the end of the prophase and which act as unit structures
during mitosis.
2. Time of Segregation. — There is a very wide-spread habit among
biologists, especially geneticists, of referring to one of the maturation
divisions as the reduction and the other as the equation division. The
facts in regard to the segregation of the sex-chromosomes in a number
of organisms have been available for years, and they stand in direct
contradiction to such views. The true status may be seen from the
following data summarized from the recent edition of E. B. Wilson's
valuable book:
326
E. ELEANOR CAROTHERS
Accessory
XY
Pre-reduction
Post-reduction
Pre-reduction
Post-reduction
Orthoptera
10 genera
1 genus
1 lomoptera
4 genera
1 Irtcroptera
o genera
4 genera
1 1 LMMKT.l
( 'uk'Optrl.i
2 genera
2 genera
6 genera
Nematoda
2 genera
2 genera
1 genus
Diptcra
3 genera
Other evidence, which has been available for years, demonstrates
that at least some of the euchromosomes behave in a similar manner.
Wenrich (1916) showed that the homologues of the tetrad which he
designated as Ci segregate half the time in the first and half the time in
second maturation division. He also showed that the homologues of
tetrad B and other combinations of tetrad C segregate uniformly in the
second division, while the writer (1913) and Robertson (1916) had
shown that segregation, in what are probably comparable tetrads of
certain other species, occurs constantly at the first division. Obvi-
ously, therefore, neither division can be referred to accurately as the
segregation (reduction) division; the term is applicable only to the
separation of the maternal and paternal components of any chromo-
some and not to either maturation division. Further evidence that the
object to be attained and not the time of attaining it is the essential
feature of maturation will be presented in this paper.
(.Quotations from a few of the best current text-books of biology will
illustrate these points. The following statement occurs in one a
propos of the first maturation division:
" But, and this is the crucial point, in the early anaphase the
members of each pair are separated, one synaptic mate going to each
pole of the spindle. Thus each of the daughter cells— SECONDARY
SI-KRMATOCYTES — receives half the total number of chromosomes that
were present in the primary spermatocyte or the somatic cells. The
essential difference between this type of mitosis (Knuri TIONAI. DIVI-
SION) and that involved in other nuclear divisions (EQUATION DIVI-
SIONS) lies in the separation of entire chromosomes (synaptic mates)
instead of the splitting of each chromosome."
This book is entitled "Foundations of Biology" and the author
states in the preface that it is intended for college students and the
general reader. Yet, among the first things that a college student in
many l.iKor.iiory courses in zoology will see for himself, is the fact that
the number of chromosomes in the first <><">ryte or spermatocyte is one-
half that in the somatic cells and is the same as the number in the
MATURATION AND SEGREGATION 327
second oocyte or spermatocyte. Such a statement, therefore, confuses
the student or else gives him an incorrect conception as to the use of the
word chromosome. Furthermore, it gives him an erroneous idea as to
the process and time of segregation, which as already mentioned is not
confined to either maturation division, exclusively.
The following quotation from another author shows improvement
over the preceding in that there is recognition of the fact that segrega-
tion may occur at either division, but a similar lack of a clear concep-
tion of a chromosome :
' Tetrads. — The pairs of chromosomes often do not appear as double
bodies; for while the chromosomes have been coming together they
may also have divided. Each pair thus consists of four half-chromo-
somes; and the quadruple body formed is called a tetrad. Owing to its
origin, two of the parts of each tetrad are maternal, the other two
paternal. In the two maturation divisions the tetrads are divided, in
two planes, first into double bodies called dyads, next into their single
components."
"First Maturation Division. — A spindle is formed, on which the
tetrads take their place. How the tetrads are divided depends on the
way they are placed on the spindle. In part, this position appears to be
fixed and always the same in the same species. In the illustration they
are represented as having been so placed that the maternal half of the
tetrad is separated from the paternal half. It is a matter of chance,
however, whether the paternal half is turned toward one end of the
spindle or toward the other. It may happen therefore, that all of the
paternal dyads go into one cell and all of the maternal dyads into the
other, or, as in the figure, part into one cell and part into the other.
The cells produced by this division are called secondary spermatocytes."
" It is worthy of note that in the division just described, no chromo-
somes have divided. The tetrads have divided, but merely by the
separation of the two chromosomes which had previously come to-
gether. Such a division is called a reductive division; it never occurs in
cell divisions except in maturation, and in only one of the maturation
divisions."
Tetrads are chromosomes according to any accepted definition of
the word; hence, "chromosomes" have divided in the above instance.
Also since a tetrad is a chromosome the statement that it is four "half-
chromosomes" is rather confusing. The difficulty disappears if the
statement is changed to "Each pair thus consists of four chromatids."
Two other points in the above quotation, while actually accurate, are
inadequate, so far as the inexperienced student is concerned. One is:
"It may happen, therefore, that all of the paternal dyads go into one
cell and all of the maternal dyads into the other." The student gains
the idea that this is a reasonably frequent occurrence, whereas, in an
organism like a short-horned grasshopper with its 24 or 23 chromo-
E. I I.EAXOR CAROTHERS
somes (12 pairs), assuming the simplest possible conditions with no
crossing-o\er. OIK- gamete in each 4096 would be expected to contain
the haploid set of 12 chromosomes contributed by a particular parent
while, in man with 48 chromosomes (24 pairs), the ratio is 1:16-
777216 (22J). The other misleading statement concerns the reduction
division: "... it never occurs in cell divisions except in maturation
and in only one of the maturation divisions." As a matter of fact, as
previously pointed out, reduction (better, segregation) occurs in both
maturation divisions; but, obviously, a given pair can segregate in only
one, though a corresponding pair in another cell may segregate in the
other. In other words, the term reduction division should only refer to
the separation of the maternal and paternal components of any chromo-
some and not to either maturation division taken as a whole.
May I present just one more quotation on this subject from another
author:
"... Weismann in 1888 prophesied that in one of the maturation
divisions it would be found that the chromosomes do not divide longi-
tudinally but transversely so that the hereditary characteristics instead
of being equally partitioned between the daughter cells would be di-
vided crosswise so that the daughter cells would receive dissimilar
groups of biophors. The ordinary longitudinal division of the chromo-
s< nnes he called an ct/n/ilion division and the extraordinary hypothetical
division during maturation the reduction division. "
'The fulfilment of this prophecy by a host of different observers \\ as
a remarkable justification of the imagination in science. The reduction
division in MHIH- form or other, often complicated and atypical, was
revealed in type after type of animals and plants until today it is gen-
erally if not quite universally accepted as a typical phenomenon of
maturation."
Of course, the last sentence may be stretched to cover a multitude
of views, but the idea which is conveyed to the student as one to be ac-
cepted without question is that there is a transverse or cross division of
chromosomes. Weismann reali/ed that reduction in number of "ids"
must occur before fertilization ; otherwise, the number would be doubled
each generation. He predicted that reduction would be found to occur
during maturation and suggested two ways in which it might be
brought about, either by a sorting out of chromosomes into two
Hinilar groups, one of which would go to each pole without division of
the constituent idants (chromosomes) or by a transverse instead of a
longitudinal division of each individual chromosome. For the intrin-
sic process, whatever the method, he proposed the term reduction
division. The ordinary longitudinal division was already known as an
divisinn.
MATURATION AND SEGREGATION
The cytologists of the time were quick to show that no transverse
division of the chromosomes occurs and that XYeismann's first sugges-
tion was the correct one. The mechanism which insures the separation
of the members of the two groups is the initial reduction in number of
chromosomes through the synapsis of longitudinally split homologues
(pseudo-reduction) followed by their actual distribution to different
cells by the two following divisions. The division which separates the
parts of any chromosome derived from one parent from those derived
from the other is the reduction division in \Veisman n's sense for that
particular pair of homologues. So much will do for the first two points,
the last remains to be considered.
3. The Historically Correct Usage of the Word Maturation as Applied
to Gametogenesis. — The early usage of the terms spermatogenesis,
oogenesis and maturation was perfectly logical and clear-cut. One
group of investigators was concerned with the origin and early history
of the germ cells and called the entire process spermatogenesis in the
male and oogenesis in the female. The other group was interested in
the ripening (maturation) of the egg and its attendant phenomena.
A brief survey of the two groups will make my point clearer. In
the first, we find v. la Valette St. George who from 1865-76 published
a series of four papers entitled, " Ueber die Genese der Samenkorper." *
Much of our present terminology on spermatogenesis was proposed by
St. George in the last of these papers. In this group fall, also, the fol-
lowing men who were largely instrumental in developing the theory of
the continuity of the germplasm; Richard Owen, 1849, called attention
to certain distinctions between body and germ cells. Virchow, 1858,
was led to enunciate his famous dictum, "Omnis cellula e cellula."
Certainly, recognition of the fact that cells arise only from preexisting
cells was an essential step in establishing the idea of the continuity of
the germplasm. Jaeger, 1878, used the expression "Continuitat des
Keimprotoplasma." Credit for establishing this theory, however,
goes to Nussbaum whose work, 1880, on the early development of the
frog and trout led him to a clear statement of the concept of the con-
tinuity of the germ cells and of the evidence for his conclusions. And
finally, Weismann, 1883, directed attention to the bearing of the con-
tinuity and comparative isolation of the germ cells on theories of evolu-
tion and heredity.
Leaving this hasty summary of the early work on spermatogenesis
and oogenesis, let us turn to the other group of investigators: who were
concerned with neither the origin nor the early development of the
germ cells but with the maturation (ripening )of the egg. (As we shall
1 This was not the end of the series.
330 E. ELEANOR CAROTHERS
see, the sperm was not supposed to undergo such a process until Van
Beneden recognized the essential feature of polar body formation to
be the elimination of chromatin.)
The gradual recognition of the essential features of the ripening of
the egg may be summarized briefly as follows: The germinal vesicle
\\ as discovered by Purkinje, 1825. A polar body was first figured,
apparently, by Carus, 1824, for a mollusk egg. Carus, however, gave
no adequate description of the structure and offered no suggestion as
to its function. Yon Baer, 1827, noted in the hen's egg the migration
of the germinal vesicle to the periphery of the yolk and its disappear-
ance. He believed both processes to be concerned with the maturation
of the egg. Dumortier, 1837, saw and described the two polar bodies
in a mollusk egg. He believed them to be the Purkinje (germinal)
vesicle. F. M tiller, 1848, suggested that these bodies were concerned
in the determination of the early cleavage planes and accordingly ap-
plied the name " Richtungsblaschen" to them. Robin, 1862, in recog-
nition ol this constant relation to the cleavage planes called them
"Globules polaires," hence our term polar bodies. Mark, 1881, was
the first to suggest that the polar bodies should be regarded as rudi-
mentary eggs. Somewhat earlier, 1875, Yan Beneden wrote that
maturation clearly consisted in the breaking down of the germinal
vesicle, the formation of the polocytes and the return of the nucleus
into the yolk. By 1883 the same author had worked out the relation
of the chromatin to polar body formation and had recognized this as
the essential feature of the ripening of the egg. He also noted the
equivalence of the male and female pronuclei in regard to amount of
chromatin. (The word chromosome was not coined until 1888.) He
then prophesied that a process whereby the amount of chromatin is
reduced would be found to occur in spermatogenesis, and later, 1887, in
collaboration with Julin demonstrated such to be the case, and that
the sperm as well as the egg underwent maturation, the essential
feature of which is chromatin reduction. Only in later text-books do
we find such a confusion of ideas as may be illustrated by this quota-
tion: "The maturation of germ cells in the male is called spermato-
genesis, in the female oogenesis." Let me repeat that in contrast with
the idea of maturation as a process concerned with the reduction in
amount of chromatin the concept of gametogenesis includes the entire
process from the time the germ cells are first recognized through multi-
plication, growth, maturation, and in the male, transformation of the
sperm.
This may seem to be an unusual introduction to a scientific paper,
luit 1 can only add that in dealing with advanced students in Zoology
MATURATION AND SEGREGATION 331
I have found their instruction faulty on the above points and believe
that the confusion is not necessary. In any case, the present paper is
concerned chiefly with the second of these points, the time of segrega-
tion. The other two are matters of definition and priority of usage.
The problem, then, is: When does segregation (reduction in Weis-
mann's sense) occur? As already shown, this question can apply only
to individual pairs of homologues and not to either maturation division.
Three conditions render the answer difficult. First, usually the homol-
ogous chromosomes are indistinguishable morphologically. Second,
parasynapsis is the method of union, at least in many forms, and is pre-
ceded by the splitting of the homologues. Finally, the four chromatids
which are parallel during part of the prophase, later form equal-armed
crosses and figures of 8 in such a manner that the chromatids which are
together in one arm or loop are separated in the next.
Wilson, McClung and Wenrich are among the few who, when con-
vinced of parasynapsis, recognized at once that, in view of the structure
of the later prophase figures, they could not determine which of the
four chromatids came from a given parent, and hence, which division
separated sister chromatids and which homologues.
We must resort then to other means for determining when segrega-
tion occurs. Information is available from four sources. (1) Sex
chromosomes, either XY pairs or the unpaired accessory chromosome;
(2) Heteromorphic homologous chromosomes; (3) Polyploidy, and (4)
Genetical evidence. The data from the first have been summarized,
the last two will be considered in the discussion, while evidence to be
presented concerns the second. WThen the homologues are unlike in
size or shape it is a simple matter to observe when segregation occurs.
II. MATERIAL AND ACKNOWLEDGMENTS
This detailed study of segregation of unequal homologues is based
on males from one species, each, of three genera of short-horned grass-
hoppers, distributed as follows: 71 Trimerotropis citrina from Kansas,
Texas and Florida, 10 Mecostethus gracilis from Maine and Michigan
and a number of Amphitornis bicolor from Kansas and Colorado. In
each of these three species certain individuals have one or both of the
two smallest pairs of chromosomes composed of homologues of different
sizes.
I am indebted to Dr. WT. R. B. Robertson for T. citrina from Law-
rence, Kansas and to Dr. H. B. Baker for M. gracilis from the Univer-
sity of Michigan Biological Station near Cheboygan, Michigan. The
work was done at the Marine Biological Station at Woods Hole and the
Zoological Laboratory of the University of Pennsylvania.
332 E. ELEANOR CAROTHERS
III. OBSERVATIONS
1. ,:!i«>i of Unequal Homologues. — (1) Trimerotropis citrina:2
.\ spermatogonial complex from a typical individual of this species
consists of 23 telomitic chromosomes which may be arranged according
to si/e in two groups. The first is composed of two small pairs. The
members of one of these pairs are about two-thirds the size of those of
the other pair. The second group consists of nine closely graded pairs
and the accessory. There is a decided break between the two groups;
as shown in Plate I, Fig. 1, the members of the smallest pair in the
second group are more than twice the length of the members of the
larger pair of the first group. In this and the following plate, the
members of the two smallest pairs are shown in solid black in order to
facilitate recognition, Figure 4 is a side view of a first spermatocyte
Irom an individual with a spermatogonial complex such as has just
been described, while Fig. 1 1 is a polar view of a second spermatocyte
with a corresponding complex.
Sixty of the seventy-one specimens studied are, chromosomally, of
tin- sort just described, but the remaining eleven (nearly 16 per cent)
have a different complex. These last have only three chromosomes in
the small group, the two members of the small pair and one normal-
-i/.ed member of the second pair (Figs. 2 and 3), but have an additional
chromosome in the second group, which conu> in the size series among
the members of the third and fourth pairs. Study of the first and sec-
mid spcniiatocytes shows the additional chromosome in this group to
be tin- s\ napiir mate of one of the small chromosomes in the first group.
The size relationship of the members of this unequal pair is sho\\ n most
strikingly in side views of second spermatocyte anaphases 'Fig. 10).
2 Unfortunately, Dr. R. I.. King, in a paper dealing \\ith 1 1 iree species of Tn'mero-
tropis, has used t lir n.une of a submenus, Pseudotrimerotrofris, as the name of the genus.
Such a procedure is not just iln-il, ,is will appear from the following stai rim-nt . McNeill
(1901), in his " Revision of the (ienus Trimerotropis," arranged the species of Trimero-
tropis in two subgenera, Agonozoaand Trimerotropis. Rrlm ' I'KM , pointed out thai
McXeill had placed the type species of Trimerotropis in his submenus Agonozoa and
used Trimerotropis s.s. for another submenus, lit- added, "It is c|iiite apparent
Agonozoa is a synonym for restricted Trimerotropis and a ne\\ name is necessary for
MeXeill's subgcnus Trimerotropis. To supply the deficiency I propose Pseiido-
trimerotropis." Kirby (1910) in his "Catalogue used all three sub^enera as genera.
< audell (1911), in a critical review of Kirby 's catalogue, states: " Pseudotrimerotropis
Rehn, of which Trimerotropis vinculata may be taken as i lie i \ pe, i> based on charac-
ters ulii<li are not, in the reviewer's opinion, of generic ini|mriaiier and the genus
should be sunk in synonymy under Trimerotropis." I lie lullnu ini; statement by
Rehn ii appendecl as a footnote to Caudell's article. " I he name Pseudotrimerotropis
\\a- |ii'i|.i>M'(| tu replace the restricted Trimerotropis <>l \b \eill, true Trimerotropis
bein y i-i|iial tn his . l^nnozoa. The author of the name has never considered it of more
than SUbgeneric rank."
MATURATION AND SEGREGATION
Lateral views of first spermatocytes show that the tetrad formed by
this pair may divide either equationally (Fig. 5) or reductionally (Fig.
6). A count of over 300 division figures in individual numbered 1571
gave 90 per cent of the former to 10 per cent of the latter. An equa-
tion division in the first spermatocyte is, of course, followed by a reduc-
tion division in the second (Fig. 9), while a reduction division in the
first entails an equation division in the derived second spermatocytes.
The last are of two sorts in regard to the pair under consideration.
Those which receive the small homologue are identical to the seconds in
typical individuals; for example, compare the two small dyads in Fig.
12, which is from specimen No. 1571 where the homologues are unequal,
with those in Fig. 11, from an individual where these homologues are
equal in size. The other sort have but one small dyad ; the second is
replaced by the large homologue as is shown in Plate II, Fig. 15. (This
figure is also from individual No. 1571.)
Another occasional condition sheds light on the variation in time of
segregation of this pair and also on one of the functions of synapsis;
namely, that it is a mechanism which insures segregation which, other-
wise, does not necessarily occur. When unequal, these homologues at
times come into the first spermatocyte metaphase as separate chromo-
somes ; whether they synapse and separate before the metaphase or do
not synapse at all is not known. A count of 52 metaphases in another
specimen (No. 1927) gave eight in this condition. Such first spermato-
cytes contain thirteen chromosomes, three of which, the accessory and
the unsynapsed, unequal homologues, are dyads. Any one of four
results may occur as a consequence of such a condition. 1. One dyad
may go to each pole undivided. This is, in effect, a segregation division
and gives second spermatocytes of the usual types. 2. Both may go to
the same pole undivided. Thus, in effect, an entire tetrad goes into
one cell while the other lacks any representative of these homologues.
All chromosomes in both cells are dyads and will divide in the ensuing
division. Inevitably, certain sperm will carry chromosome number 2
in duplicate and should any of these fertilize an egg, the resultant
zygote would necessarily be triplicate in respect to the factors carried
by this chromosome. 3. Both may divide as shown in Fig. 7. This is
in effect an equational division, but each of the resulting second sperma-
tocytes receives two monads (Figs. 13 and 14) which are incapable of
division at this time; one monad may pass into each spermatid, both of
which would consequently contain a full complement of chromosomes,
or both monads may pass into the same spermatid. The result in the
last case is the same as in "2," above. 4. One may divide while the
other passes undivided to one pole. Figure 8 shows the larger dyad in
334 E. ELEANOR .CAROTHERS
an equational division, while its small homologue is going undivided to
one pole. One of the four spermatozoa derived from such a first
spermatocyte would again contain this chromosome in duplicate; one
\\ould lack it completely; and two would be normal. All of the de-
ibed conditions have been found except the converse of that shown
in Fig. 8, where the small dyad divides and the larger one segregated,
and I have little doubt that such complexes exist, also. One should
bear in mind that while a tetrad can undergo two divisions without any
EXPLANATION OF PI.ATKS
The complexes were drawn with the aid of a camera lucida at a magnification of
2800 diameters. They were reduced ^ in reproduction.
The chromosomes under especial consideration on each plate are in solid black.
All lateral views of entire complexes, except Fig. 12, are from two and sometimes
three sections.
EXPLANATION OF PLATE I
(Trimerotropis citrina, entire complexes)
FIG. 1. Spermatogonium, polar view, too small pairs, 19 large chromosomes,
(9 pairs and the accessory).
FIG. 2. Spermatogonium, polar view, one small pair. The synaptic mate of
the third small chromosome is one of the 20 large chromosomes of which there are 9
pairs, the accessory, A', which is unpaired, and one medium-sized unpaired chromo-
some. The chromosome set off by the dashed line is in the adjoining section but is
obviously the homologue of the similar one near it.
FIG. 3. Similar to the last (all chromosomes in one section).
FIG. 4. First spermatocyte, side view, homologues equal; 60 of the 71 indi-
viduals studied were of this type. Figures 1 and 11 illustrate the spermatogonia and
second spermatocytes, respectively, which are characteristic for these specimens.
FIG. 5. First spermatocyte, side view, homologues of one small pair unequal;
from same specimen as Spermatogonium shown in Fig. 3. Fk-viTi out of 71 indi-
viduals are of this type. I )ivision of unequal pair equational.
In,. 6. Similar to Fig. 5. Division of unequal pair reductional.
FIG. 7. First spermatocyte anaphase. Unequal homologues not synapsed.
Both dyads dividing, in effect an equational division.
F'IG. 8. First spermatocyte anaphase from same specimen as Fig. 6. Dyads of
unequal pair not synapsed; The larger dyad dividing equationally, the smaller
segregating.
FIG. 9. Second spermatocyte, polar view, derived from a first spermatocyte
division such as is represented by Fig. 5.
FIG. 10. Same type as preceding, side vit-u .
I'n.. 1 1. Second spermatocyte, polar view, from nm- of i In- (>() t ypir.d sprrimi'ns.
I id. 12. Second spermatocyte, side view, composition of small dyads identical
to those in preceding cell but derived through segregation of unequal homologues in
the first spermatocyte. Drawings from same individual as Figs. 2 and 15. This cell
received the smaller homologue, that shown as Fig. 15 the larger one.
Fi<;. 13. Second spermatocyte, polar view, nypr with accessory). Thirteen
chromosomes, two of which are monads, derived from such a first spermatocyte as
that shown in Fig. 7.
FIG. 14. Second spermatocyte (type lacking accessory), side view, entire com-
plex. Unequal homologues not synapsed. Both monads derived from such a first
spermatocyte as is shown in Fig. 7.
MATURATION AXD SEGREGATION
335
6
PLATE I
1)
f5
^
c2_ o
<?
^b
^1
^
'£>
^~? %^
^7'/^'^5
rcO
9
8
'/?
10
OQ
14
12
336 E. ELEAXOR CAROTHERS
reconstruction, a dyad can divide only once, and a monad
cannot divide at all but must simply pass to one pole.
In three of the 71 specimens of this species studied the smallest
pair also is composed of unequal homologues. But, the members of
this pair ordinarily segregate in the first maturation division (roughly
(>r< per cent of the observed instances) in contrast to about 90 per
cent post-reduction shown by the pair just described. In this respect
the smallest pair approaches the unequal pair in Bracliystoln ma«nti
(Carothers, 1913).
Figures 16 to 19, inclusive, are from an individual in which the
members of both small pairs are unequal. Figures 16 and 17 are in-
complete lateral views of first spermatocyte complexes. The first
shows both unequal tetrads dividing equationally, the latter the smaller
pair dividing reductionally. Figures 18 and 19 are from second
spermatocytes and show the segregation which follows an equational
division of the first spermatocyte, such as is represented in Fig. 16.
In case segregation occurs in the first division, as illustrated in Fig. 1 7,
the second spermatocytes show a simple equational division of dyads
with like chromatids.
EXPLANATION OF PLATE II
IK;. 15. Second spermatocyte, polar view, only one small dyad. The homo-
logues of the unequal tetrad segregated in the preceding division and this cell re-
ceived the larger component.
I n,. in. Part of first spermatocyte complex, side view, showing both small
tetrads unequal .UK! dividing equationally.
I IG. 17. From same individual. Smallest tetrad dividing reductionally.
FIG. 18. Part of second spermatocyte metaphase, side view, members of small-
est pair dividing reductionally.
In,. \(). Similar to preceding, both small pairs dividing reductionally.
1 n,. _'0. Lateral view, first sperm, iin< \ ie anaphase. Mi'n'stftlinx gnicilis.
I lomologuesof both small p.iirs unequal. The members of one pair have segregated;
those of the other pair have divided equationally.
1 n.. 21. Slightly earlier stag*- than tin- preceding, both of the unequal tetrads
dividing equal ionally.
In.. 22. Second spermatocyte, polar view, same species. Segregation division
for unequal homologues of both small pairs.
I ii,. 23. I ir^i -|"-i m.itocyte anaphase, side view, Ani(>liitnrnis bi color, both
Miiall tetrads unequal and dividing equationally.
I IG. 21. Second spcrmatocyte, side view, from same species showing reduc-
i ioii.il ili\ isinn of i he i uo small pairs.
I IG. 25. Sister second spermatocyte metaphases, A»i/>hitoniis bicolor, side
\ ie\\, one dyad caught in plate at preceding division. < >nc monad free in each cell.
I IG. 26. Unequal tetrads in late prophase, T.dtritm. .\e\\l\ acquired portion
more dense.
FlG. 27. San ir in MrnntrtliHS gracilis. Note the tendency for the newly ac-
quired portion to organize itself into chromomeres.
I IG. 2,X. I .at i- | n 1. 1 ih. isc ten. ids T. citrina, showing chromomere vesicles.
MATURATION AND SEGREGATION
337
PLATE II
17
23
25
18
Ji
26
_
27
19
28
e
22
23
J3S E. ELEANOR CAROTHERS
Let us consider briefly the other two species. (2) Mecostethus
.v/7/.v: This species along with others of this genus has been studied
in detail by Prof. C. E. McClung and I shall refer to it only in regard to
the point in question.
The first five males to be added to our collection were taken at
Salisbury Cove, Maine, in the summer of 1923. All show the second
pair unequal and two out of the five have both small pairs unequal.
The specimens taken nearCheboygan, Michigan, are of especial interest,
for while some of them are of the types just mentioned, others supply
a link which although not unknown in other species, has been lacking
hitherto for both .17. gracilis and T. citrintr, namely, they are of the
theoretically expected type with the second tetrad composed of two
large homologues. While this condition automatically throws the te-
trad out of its usual place in the size series, it is still readily recognized
by certain peculiarities.
As to time of segregation, the homologues of both pairs when un-
equal may undergo either pre- or post-reduction. A count of 150 cells
from three consistent individuals gave a ratio of pre-reduction to post-
reduction for the second pair of 1:8, while the ratio for the first pair
derived from a count in 63 cells was 1:12, which is not very different
from that for the second pair. The numbers were not extended be-
cause counts from another specimen demonstrated that the range of
variation from individual to individual is such that ratios may have no
significance. Both small tetrads in this specimen are unequal and no
attempt was made to distinguish between them further than to ascer-
tain that both varied as to time of segregation. A count in 55 cells
u.i\f a combined ratio of approximately 1:1. Kvidently either one
or both of these tetrads in this individual behaves differently from
either of them in the other three individuals. A closer study might
reveal a corresponding morphological variation such as Wenrich
(1916) found for tetrad C in Phrynotettix. Drawings from .17. gracilis
are shown on Plate II. Figure 20 is from a side view of a first sperma-
tocyte anaphase. One pair was dividing equationally, the other reduc-
tionally. Figure 21 is a similar view from a slightly earlier stage in
which both pairs were dividing equationally. Figure 22 is a polar view
of a second spermatocyte metaphase illustrating post-reduction for
both pairs.
(3) Amphitornis bicolor: Individuals of this species also may have
either one or both of the two smallest pairs unequal. Our material
i- interesting for several reasons: in the first place certain individuals
were rolln it-d I >\ 1 )r. \V. S. Sutton at Russell, Kansas, over 30 years
ago, others were taken by myself in 1919 and 1921 at three widely
MATURATION AND SEGREGATION 339
separated localities in Kansas and in the vicinity of Pike's Peak in
Colorado, so that the range of this material both in area and time is
considerable.
The unequal homologues in this species show a much greater pre-
disposition than in either of the others towards post-reduction. Pre-
reduction occurs in not over 1 per cent of the cases noted.
Figure 23 is a lateral view of a first spermatocyte anaphase and
Fig. 24 a similar view of a second spermatocyte metaphase; both show
post-reduction for these two pairs.
2. Chromomere Vesicles and the Origin of These Unequal Homo-
logues.— Most cytologists who have studied unequal homologues have
reported, without very satisfactory evidence, however, that the in-
equality was due to a loss of chromatin by one member of the pair.
This doubtless is the cause sometimes, but I believe that in the three
species concerned in the present paper all of the inequalities described
are due to increases; and, furthermore, these increases are not fictitious
ones due to chromatin which belongs to some other chromosome having
been acquired by these enlarged homologues. I am convinced that the
increase is brought about by the transformation of terminal granules
into chromomere vesicles which become densely chromatic and are
maintained. An idea of the process may be gained from Plate II, Figs.
26, 27 and 28. Vesicles such as are shown in the last figure take a dense
chromatic stain in earlier stages when the chromatin threads stain very
lightly. Gradually, as the threads take a more dense stain the vesicles
become less densely stained until at one point the vesicles are seen to be
filled with a flocculent chromatin which is the same in appearance,
practically, as that in the chromatin thread. (These observations
have been verified by the use of Feulgen's nucleal reaction which gives
results similar to iron-haematoxylin for chromomere vesicles.) Ordi-
narily, the vesicles continue to lose their staining capacity until they
become indistinguishable. That they really may be persistent struc-
tures, however, is indicated by the fact that they have been recognized
in both metaphases and early anaphases. Certainly, in spermatogene-
sis they are always associated in the prophases with definite chromo-
meres of particular chromosomes (Carothers, 1913, p. 498). Wenrich
was the first to note that these vesicles may be terminal instead of
subterminal. He wrote (1916, p. 113), "Are polar granules to be
classed in the same category as the plasmosomes (chromomere ves-
icles)? Is it possible for a polar granule to become transformed into a
plasmosome and then back into a polar granule again? The last ques-
tion seems to be answered in the affirmative by the conditions in
Phrynotettix. In the case of B, for example, one of the proximal
340 E. ELEANOR CAROTHERS
granules becomes "expanded" in only about 16% of the cases counted.
In becoming expanded it has become a plasmosome. When it is not
expanded, it remains a polar granule." \Yenrich regarded the larger
homologue as the normal type and believed that a loss of chromatin
had occurred and, hence, did not suspect that this "expanded" condi-
tion might lead to a permanent enlargement. I believe this, however,
to be the logical conclusion, since growth of chromosomes involves a
similar mechanism. After each cell division, the chromosomes become
diffuse or "expanded"; in many instances, they form chromosomal
vesicles; when they again condense, they have grown back to a size
characteristic for each ; that is, they have doubled their mass during the
period of diffusion and are now ready for another division.3
\Yhatever the function of chromomere vesicles, granting, as seems
likely, that they have some specific function, the associated effect is,
in some instances, a prolonged period of diffusion (e.g., Phrynotettix}
which may well lead to a differential increase of chromatin for the
chromomeres concerned.4 The same mechanism may function equally
well to secure reduction in size of chromomeres if the period of conden-
sation is prolonged and the period for growth (expansion) consequently
shortened or, as is usually the case, the periods of condensation and
diffusion may be so balanced as to maintain a constant size. Figure 27
from .17. vnicilis is of particular interest as it indicates that these en-
larged terminal vesicles may be organized finally into chromomeres.
Sec also McClung, 1928. Plate XXVIII, Figs. 5\d, 53 and 56. As
already mentioned, in certain members of this species from Michigan
both homologues of the second pair are of the large type. The tetrad
formed by such a pair loses its usual position in the size series but is still
recognizable as it forms a ring instead of the V which is characteristic
for tetrads of this genus.
If my suggestion is correct, then new chromatin has been organized
to form a permanent component of the complex; and, granted that the
chromatin bears hereditary factors, a mechanism for progressive
changes is shown. In other words, if the chromomere vesicles charac-
teristic of certain chromosomes are capable of a permanent modification
in kind or amount of chromatin in response to a changed environment
with it> resultant need for an altered metabolism, these changes would
In- adapt i\ i- in character.
3 \V. K. de Mol (1927), in reference to Hyacinthus oricntalis, states that the chro-
mosomes probably receive material from nucleoli either at permanent secondary con-
strictions or by satellites which are really nucleolar globules not taken up in the
chromosomes.
4 Gertrand Hasse-Hessell (1928) believes that the satellites are differentiated
portions of chromosomes, organs for assimilation of chromatin, and that the satellite
is the morphological expression of chromatin absorption on its active surface.
MATURATION AND SEGREGATION 341
Attention should be called to the fact that in the Orthoptera the
only homologous chromosomes which have been found to be unequal in
size are those which form the two or three smallest tetrads and these
are precisely the ones which are usually characterized by chromomere
vesicles.
3. An Accident During Mitosis, A. bicolor. — Ten or twelve second
spermatocytes in one individual show a dyad caught in the "zwischen-
korper," resulting from the first spermatocyte division, with one
chromatid in each daughter second spermatocyte (Plate II, Fig. 25).
The abnormality is due, apparently, to some slightly pathological state
which existed at the time of the first spermatocyte divisions and seems
worthy of record as a morphological condition which may explain oc-
casional aberrant genetical ratios. My expectation was that the
trapped dyad in each case would be found to belong in one or the other
of the second spermatocytes concerned, and that one of them would
possess a normal dyad derived from the same tetrad. Drawings of
entire complexes show, however, that each of the affected seconds actu-
ally have their normal number of chromatids, since there is a monad in
each equatorial plate which, together with the monad of the trapped
dyad in the corresponding cell, accounts for the components of the
expected dyad. Figure 25 shows the chromosomes of two second
spermatocytes derived from one such first. The upper complex con-
tains eleven dyads including the accessory, which is easily recognized
by its roughened outline and one monad. The lower complex contains
ten dyads and one monad. A small segment of the boundary between
the two cells is represented together with the trapped dyad. In the
drawings the two equatorial plates are placed nearer together and con-
sequently nearer the median cell boundary than they actually are for
the sake of economy of space on the plate.
Stages later than that shown were not available, but the monads in
the equatorial plates would behave, doubtless, as other monads in
similar situations; that is, each would pass to one pole without division
so that of the four spermatids derived from these two cells, two would
be normal and the nuclei of the other two would lack one chromosome
with the result that an egg fertilized by one of the latter would be hap-
loid for this chromosome. The additional possibility exists that the
monad left at the cell boundary may form a chromosomal vesicle and
migrate to join with the pronuclei in fertilization, in which case the
production of individuals triploid for this chromosome would follow
fertilization by any sperm which received both monads.
4. Incipient Octad Formation — T. citrina, individual No. 4853.—
This specimen, in addition to being one of the eleven T. citrina which
342 E. ELEANOR CAROTHERS
have an unequal pair of homologues, exhibits a tendency on the part of
t\\o non-homologous pairs to form a multiple. On account of the
latter peculiarity, this specimen yields interesting evidence on three
points: il) an unusual mode of octad formation; (2) the behavior of
spindle fiber attachments; and (3) the time of segregation. The indi-
vidual is one of fifteen taken at Kingman, Kansas, and is the only one of
the 71 T. citriun studied which shows any tendency toward multiple
formation. The insect seems to be typical externally and the testes
are normal.
The occurrence of octad multiples was predicted by McClung
(1905). They were first recognized by \Yoolsey (1915) in Jamaicdini,
and Robertson (1916) correctly suggested that the three atelomitic
rings in the first spermatocytes of Stenobothrus and Chloi'ultis were
octads. They have since been found in Ucsperotettix, Circotettix,
Stauroderus, and a subtropical genus, Sphenarinm. The number of
multiples is constant for species and perhaps even genera in Stenobo-
thrus, Chlncaltis and Circotcttix. McClung (1917) has shown that, on
the other hand, in Hesperotettix viridis there is considerable variation
within the species, but constancy for any given individual. The
specimen of T. citrina under consideration is unique in that it shows
variation from cell to cell within the individual, ranging in the first
spermatocytes from no multiple through an octad with one homologue
of each tetrad united by their distal ends to those where both pairs are
so associated. This attachment at the distal ends is another point of
difference between this multiple and those previously described where
the union occurs at the ends associated with the spindle fibers.
The two pairs concerned are of intermediate size, one somewhat
larger than the other, as may be seen in the spermatogonium figured
(Plate III, Hg. 29).5 Here two non-homologous chromosomes (solid
black) are united a! their distal ends. ( )ne is perhaps one-fourth longer
than the other. The homologue of each is free and indistinguishable
among the other members of the complex. This is the only clear
spermatogonial complex which was obtained, though some cells seem
to show the full complex of 23 chromosomes and others to have t\\o
multiples resulting in 21 chromoMunt-.
The first spermatocytes show much variation in the behavior of
these two tetrads. The more usual condition is for them to form a
ring-shaped octad (Plate III, Figs. 30 and 2>5d). Its structure is easily
understood by comparison with the two tetrads represented in solid
color in Fig. 31. This is a drawing of one < >! the few first spermatocytes
6 These unequal homologues segregate in cither division as in the preceding
instances. '1 hry an- not inked in solid in this case as the peculiar behavior of the
tuo pairs concerned in octad formation is emphusi/ed in this manner.
MATURATION AND SEGREGATION 343
where no octad has been formed. If the free ends (the ends not asso-
ciated with the spindle fibers) of these tetrads are joined and the fiber
attachments remain as they are in the separate tetrads, the result is
the octad ring.
Morphologically, this ring differs widely from the octads of Steno-
bothrus, Hesperotettix and Chloealtis, in which the open part of the ring
lies in the plane of the spindle and where there are but two places for
fiber attachments. In the present instance the open part of the ring
lies in the plane of the equator and there are four loci for fiber attach-
ments (two for each of the component tetrads). In Circotettix, as I
pointed out in 1921, the octad often is indistinguishable from the
atelomitic tetrad. As to the origin of these three different types of
rings; the atelomitic tetrads have come about, apparently, from a shift
of fiber attachment. The octads of Circotettix, Hesperotettix, and prob-
ably those of Stenobothrus and Chloealtis have been formed by union at
the ends to which the fibers attach, while in this individual the union
has occurred at the distal ends, and fiber attachments have remained
constant though in certain cases those on one of the tetrads do not
function.
This brings up the other forms the octad may assume. Next to the
ring the most common form is that where the tetrads are united by one
arm only as seen in Figs. 32, 35a, 35c, 3>5e and 36. Such forms are the
logical descendants of spermatogonia similar to the one represented in
Fig. 29 where one homologue of each pair is free. For this type one
can say with assurance that segregation occurs in the second division.
See Figs. 39 and 43, second spermatocyte anaphases, where an inverted
V comparable to the V in the spermatogonium, Fig. 29, has segregated
from the two free homologues which were not identified in the sperma-
togonia. This is the only form which this multiple assumes where one
can be certain as to which is the segregation division. One who did
not know the synaptic relations might conclude that whatever type of
division was occurring in the multiples represented in Figs. 33 and 34,
the opposite type was represented in Figs. 32, 35a-e, and 36, but this
assumption is not justified because the chromatids of the individual
tetrads have been through a period of parasynapsis; consequently, it is
not possible to say which chromatids are derived from different
parents.
The third form in point of frequency of occurrence is like the first
in that the homologues of the two tetrads are united at the ends which
ordinarily would be free, but the dyads of the smaller tetrad are not
united with each other (Figs. 33, 34 and 356). As to the origin of such
forms, the synaptic ends of the small pair may be unable sometimes to
344 E. ELEANOR CAROTHERS
get together as a result of their position at the ends of the longer pair
so that what synapsis occurs is merely a continuation of that of the
latter and must proceed in a direction reverse to normal. This, I am
inclined to believe, is the explanation of such forms; although they may
be due to a separation of the smaller homologues after synapsis. The
end result is the same in either case. The fibers on either the larger or
smaller (but not on both) of the tetrads function in such octads. Fig-
ure 33 is of interest as evidence of a struggle for supremacy between the
libers to the two tetrads. \Yhile those of the smaller tetrad have
gained the ascendancy, there is a distinct torsion of the arms of the
larger tetrad; indeed, from this figure there might be doubt as to
whether the long arms would not yet swing into the equator and the
division occur in the opposite plane. The similar octad shown in Fig.
-U, however, has reached a stage where there is no reasonable doubt
that the division will occur in the plane indicated in Fig. 33.
A fourth type (Fig. 35/), although found only once, is of especial
interest, since such a division would give an entire tetrad to each
EXPLANATION OF PLATE III
Trlmerotropis citrina, specimen Xo. 4853
(Multiple solid black)
FIG. 29. Spermatogonial complex, polar view. Two non-homologous chromo-
somes united at their distal ends.
FIG. 30. First spermatocyte, polar view, octad multiple in form of ring.
l'it.. 31. First spermatocyte, polar view, 12 chromosomes. Components of
octad ring shown in 30 appear as two separate tetrads.
I i>.. 3J. 1 irst spermatocyte, side view, members of octad united by only one
arm of each. Entire complex not shown.
FIG. 33. First spermatocyte, side view, not complete. Members of octad
united by both arms, dyads of smaller tetrad separated but retaining their fiber
attachments.
FIG. 34. Similar to preceding.
FIG. 35. Various forms which the octad assumes in the first spermatocytes.
Fig. 35/ would result in segregation of entire tetrads.
FIG. 36. Similar to Fig. 32.
FIG. 37. Second spermatocyte, polar view, entire complex. Multiple such as
would be derived from octads shown in Figs. 33 and 34.
Fi<;. 3X. Second spennatocyte, polar view, cnl ire complex, derived from such a
first spermatocyte as is shown in Fig. 31.
I •!«.. 3'*. Second spermatocyte, side view of anaphase. entire complex. Mul-
tiple derived from forms like those shown at Figs. 32, 35<;, 35c and 36.
FIG. 40. Polar view of such a second spei m.in» \ u multiple in metaphase.
I n.. 41. Second spermatocyte, polar view, multiple derived from ring octads
like those shown in Figs. 30 and 35d.
1 n.. 42. Second spermatocyte metaphase, oblique view, multiple, a modifica-
tion of type shown in preceding figure.
I I*.. 1 v Knt ire complex similar to that shown in Fig. 39 except that this one
contains the accessory while the other lacks it.
l-ii,. II. I'arii.il complex similar to one shown in Fig. 41.
MATURATION AND SEGREGATION
345
PLATE III
29
32
37
39
42
30
33
wbw c d e • f
35
0
38
40
•/.
31
34
36
(5
(D
41
44
346 E. ELEANOR CAROTHERS
derived second spermatocyte while all four of the resulting spermatozoa
would lack one member of the normal haploid series and be duplex for
.mother. If such spermatozoa should fertilize eggs, the new indi-
viduals would be haploid for one member of the series and triploid for
another, while the number of chromosomes would remain normal.
The structure of the octad may be understood by assuming that an
octad similar to the one illustrated in Fig. 30 rotated 90° about an
axis passing through the points of union of the two tetrads in a plane
\ertical to the equatorial plate. All of the normal spindle liber at-
tachments were functioning.
In the second spermatocytes, one finds an unusual series of forms
which would be very puzzling without a knowledge of the variations
occurring in the first spermatocytes. Figure 38 illustrates the ordinary
12-chromosome form which results as one of the daughter cells from the
division of such a first spermatocyte as that represented in Fig. 31.
The form of multiple which is shown in Fig. 37 may be derived from a
first spermatocyte multiple like that in Fig. 35b with only the fibers to
the larger member acting in this division as in the first, or from those
of the type figured in 33 and 34 with the fibers to the larger component,
which were in abeyance at the first division, functioning. In any case,
half of the normal spindle fibers have failed to operate, in counter-
distinction to the remaining instances where, in both divisions, all of
the fibers normal for the two separate tetrads have functioned, although
half of them would have been sufficient for the necessary distribution ot
the parts of the octad.
Figures 39 and 43 show entire complexes in anaphase; one lacks,
the other possesses, the accessory. The multiple in both is of the type
which results from an association of only two dyads, one from each
tetrad, giving an inverted V to one pole and two free rods to the other.
The appearance of such a multiple in metaphase is shown in Fig. 40.
Figures 41 and 44 show the multiple derived from first spermatocytes
like those illustrated in Figs. 30 and 35J where both ends are united.
The result, in this division, is an inverted V to each pole.
To summarize:
1 . This multiple is not constant for the individual. \Yhen present,
it is formed by union of the distal ends of the chromosomes, in this
respect rr-einl ling certain Oenothcm multiples rather than other
( >i lliopteran octads.
2. Ordinarily each chromatid of a chromosome has a functional
spindle fiber. In this octad, instances occur where only half of them
function, e.g., Figs. 33, 34 and 37. Figure 33 indicates competition
between the two sets.
MATURATION AND SEGREGATION 347
3. Second spermatocytes show clearly both pre-reduction, Fig. 37,
and post-reduction, Figs. 39 and 43.
IV. DISCUSSION
Let us consider the four sources of information as to time of segrega-
tion mentioned at the end of section I. (1) Sex chromosomes, as al-
ready shown, may undergo either pre- or post-reduction, but one or the
other is constant for any particular species.
(2) Heteromorphic Homologous Chromosomes. — So far as the unequal
tetrads are concerned, similar conditions exist in other species in our
collection, but the observations here presented together with those pre-
viously published, Carothers, 1913; Wenrich, 1916; Robertson, 1916,
are believed to be sufficient to demonstrate the range from 100 per cent
pre-reduction to 100 per cent post-reduction. In heteromorphic pairs
(Carothers, 1916, 1921), where the difference is associated with spindle
fiber insertion, segregation has been found to occur only at the first
maturation division. Probably the reason for'this is to be found in the
mechanical conditions involved. Certainly, we should not assume
from this that when these tetrads are composed of homomorphic dyads
they also segregate at the first division.6
(3) Polyploidy. — In (Enothera and Datura mutants where the nor-
mal complex has become unbalanced, the extra dyads are reported to
segregate at the first maturation division. But to demonstrate that
such behavior is invariable would require a very detailed examination.
On the other hand, Lesley and Frost, 1928, reported, "additional chro-
mosome fragments," in two extreme "small " Matthiola plants. In both
plants these "fragments" (supernumaries?) segregate at either divi-
sion. This behavior agrees with that of the unsynapsed dyads reported
in this paper which are shown also to segregate at either division.
(4) Genetical Evidence. — Allen, 1924, from a study of inheritance of
non-sex-linked characters in the four clones of Sphcerocarpos derived
from one pollen mother cell concluded that, "in some way qualitative
segregation can be brought about in both divisions."
Whiting, 1924, concluded that in the parasitic wasp, Habrabracon
(Hadrobracon) , "The first maturation division of the egg may be either
equational or reductional for various loci apparently according to
chance." His data was obtained from females heterozygous at four
loci.
Similarly, Goldschmidt and Katsuki, 1928, in a combined cytologi-
cal and genetical study of a mosaic gynandromorph strain of Bombyx
6 For additional discussion of segregation of sex-chromosomes and heteromorphic
homologues, see Carothers, 1926.
348 E. ELEANOR CAROTHERS
mori, showed that a non-sex-linked recessive gene for skin transparency
may segregate at either maturation division.
Briefly, then, both cytological and genetical data justify the follow-
ing conclusion: Reduction in number of chromosomes should not be
confused with the segregation (reduction) division which applies only to
individual pairs. Reduction in number of chromosomes is brought
about by synapsis, while segregation of the parts of the tetrads derived
from one parent from those derived from the other results from the
two maturation divisions which follow each other in rapid succession
and together separate the four chromatids of each tetrad into different
cells.
LITERATURE LIST
References to the earlier papers may be found in the excellent bibliographies in
both E. B. Wilson's, The Cell in Development and Heredity, MucMillan, 1925, and
in E. L. Mark's paper listed below.
AI.LEX, C. E., 1924. Inheritance by Tetrad Sibs in Sphxrocarpos. Proc. Am. Phil.
Soc., 63: 222.
BELLING, J., AND A. F. BLAKESLEE, 1922. The Assortment of Chromosomes in
Triploid Daturas. Am. .\<it., 56: 339.
CAROTHKK>, 1C. ELEANOR, 1913. The Mendelian Ratio in Relation to Certain
Orthopteran Chromosomes. Jour. Morph., 24: 487.
CAROTHERS, 1C. ELKANOR, 1921. Genetical Behavior of Heteromorphic Homologous
Chromosomes of Circotettix (Orthoptera). Jour. Morph., 35: 457.
CAROTHERS, E. ELEANOR, 1926. The Maturation Divisions in Relation to the Segre-
gation of Homologous Chromosomes. Quart. Rev. Biol., 1 (3): 419.
CAUDELL, A. X., 1911. Some Remarks on Kirby's Synonymic Catalogue of Orthop-
tera, Vol. 3. Ent. News, 22: 158.
GOLDSCHMIDT, R., AND K. l\AisrKi, 1928. Cytologie des erblichen Gynandro-
morphismus von Bombyx mori L. Biol. Central., 48: 685.
KIM., K. I.., 1923. Heteromorphic Homologous Chromosomes in Three Species of
Pseudotrimerotropis (Orthoptera: Acrididae). Jour. Morph., 38: 19.
KIRBY, \V. F., 1910. A Synonymic Catalogue of Orthoptera. 3: British Mus. Nat.
Hist.
LESLEY, MAI«,\KKT M., AND HOWARD B. FROST, 1('2S. Two extreme "Small"
Matthiula Plants etc. Am. A',//., 62: 22.
MARK, E. L., 1881. Maturation, Fecundation and Segmentation of Limax campes-
tris, Binney. Hull. M/is. C</»ifxir. /.ool., Harvard Coll., 6: 173.
McCLUNG, C. 1C., 1905. The Chromosome Complex of Orthopteran Spermatocytes.
Biol. Hull., Q: M>\.
MI ( '1.1 M., ( '. 1C., 1(M7. The Multiple ( hromosomes of Hesperotettix and Mermiria
(Orthoptera'. Join. Morph., 29: 51''.
.M«('i.r\(,, (". 1C., l'^2S. Differential Chromosomes of Mecostethus gracilis. Zeit-
schr. f. /.fll. u. Mikr. Anal-., 7: 756.
M< NI.III. [., I'M)]. Revision of the Orthopteran Genus Trimerotropis. Proc. U.S.
.\<it. Mus., 23: 393.
RI.IIN, J. A. i .., T'Ol. Random Notes on North American Orthoptera. Trans. Am.
J-'.nt. SOC., 27: 334.
RoiU'.k isov, \V. R. I',., 1916. Chromosome Studies, I. Jour. Morph., 27: 179.
ST. <.I,OK<.I •;, v. i. \ VALE i n •:. 1865. t'ber die Genese der Samenkorper, Erste Mit-
tlu-il. Sihultz Arch., 1: 403.
ST. GEORGE, V. i \ VALLETTE, 1876. Cber die Genese der Samenkorper, Vierte
Mittheil. S,hnllz. Arch., 12: 797.
MATURATION AND SEGREGATION 349
VAN BENEDEN, E., 1875. La maturation de 1'oeuf, la fecondation et les premieres
phases du developpment embryonnaire des manniferes etc. Bull. Acad.
Roy. de Belgique, 44: 686.
VAN BENEDEN, E., 1883. Recherches sur la maturation de 1'oeuf et la fecondation.
Arch, de Biol, 4: 265.
VAN BENEDEN AND CH. JULIN, 1884. La Spermatogenese chez 1'Ascaride Megalo-
cephale. Bull. Acad. roy. des Sci. des Lettres et des Beaux- Arts de Belgique,
7:312.
WALDEYER, W., 1888. Ueber Karyokinese und ihre Beziehungen zu den Befrucht-
ungsvorgangen. Arch. mikr. Anat., 32: 1.
WENRICH, D. H., 1916. The Spermatogenesis of Phrynotettix magnus, etc. Bull.
Mus. Compar. Zool., Harvard Coll., 60: 55.
WHITING, P. W.f 1924. Some Anomalies in Habrobracon and their Bearing on Matu-
ration, Fertilization and Cleavage. Anat. Rec., 29: 146.
WOOLSEY, CARRIE I., 1915. Linkage of Chromosomes Correlated with Reduction in
Numbers etc. Biol. Bull., 28: 163.
Till-: SI7K OF TUF. I'.ODY AXD THE SIZE OF THE
ENVIRON MKXT TX THE GROWTH OF
TADPOLES
EDWARD F. ADOLPH
(From flic Physioloc/ica! Laboratory, The Uiiirersity of Rochester S-chool of
Mc'licine and Dentistry. Rochester, N. Y.)
INTRODUCTION
Body size is influenced to varying- degrees by environmental condi-
tions in the several phyla of organisms. Among aquatic animals, in
general, it has long been known that adult size and rate of growth are
functions of the volume of the water available as well as of the more
usually limiting environmental factors of food supply and temperature.
In the mammals it is generally assumed that the average rate of growth
is the optimal rate of growth, and that mean adult body size represents
the product of internal regulatory functions. But in poikilothermic
animals it is apparent that body size is equally controlled by conditions.
The readiness with which the body size of these animals responds to
external factors presents an opportunity for experimental analysis of
some of the size regulators.
Xumerous qualitative observations upon the growth of tadpoles in
crowded and uncrowded situations have indicated that there exist
marked effects of the size of the environment upon body size at any
given age (Pfliigcr, 1883; Yung, 1885). The natural assumption was
that the effects were due to deleterious substances accumulating in the
medium. But the experiments of the above investigators, and of Bilski
(1921) and Goetsch (1924), threw grave doubts upon this view, and
indicated that a truly spatial influence was at work. A chemical influ-
ence seemed merely pathological, but a physical influence seemed worthy
of experimental analysis. It was for the purpose of discovering the
nature of this spatial factor and its quantitative effectiveness that the
following measurements were undertaken.
METHODS
For the objective in mind, the best criterion of body size was the
body weight. This was measured by weighing one or more individuals
in a manner which gave strictly comparable results. All weighings were
done in duplicate.
350
GROWTH OF TADPOLES AND CROWDING 351
Tadpoles above 100 milligrams in weight were separated from the
water by pouring the latter into a sieve. The food material and the
debris were then allowed to remain in the sieve while each tadpole was
picked up with a perforated spoon and transferred to a dish of clean
water. All the tadpoles were then sieved together from the clean water
and emptied onto filter paper, from which they were poured into a tared
weighing bottle.
Smaller tadpoles were usually too delicate to endure the draining on
filter paper. They were poured into a tared Gooch crucible, the crucible
was drained and wiped thoroughly and then put into the weighing bottle.
For tadpoles less than 20 milligrams each in weight, it was necessary to
have in addition a tared amount of water within the weighing bottle.
For duplicate weighings the draining and taring were repeated. When
the tadpoles were drained in a Gooch crucible, more water clung to the
tadpoles than when the tadpoles were drained on filter paper. A small
correction for this water could be made by inference from older tadpoles
that could be weighed both with and without the crucible. Cultures of
less than four individuals could be weighed with sufficient accuracy only
after sizes of 40 milligrams or more had been attained.
For embryos not yet hatched from the gelatinous membranes, body
volumes were estimated from microscopic micrometer measurements of
two diameters. The specific gravity of the embryos was assumed to be
1.04; Bialaszewicz (1908) found that the specific gravity gradually
changed from 1.08 to 1.03 during the development of frog eggs up to
the time of hatching.
The plan of the experiments was to keep the tadpoles under as uni-
form conditions as possible. All eggs in comparable experiments were
derived from a single clutch or brood, and all clutches of Rana pipiens
were collected in one small stream. It is believed that paternity as
well as maternity is uniform within each clutch. Tap water was the
medium used throughout the experiments, the water usually being al-
lowed to stand in large bottles before use, so that when used all com-
parable cultures received uniform samples of water at the proper tem-
perature. Food, consisting of Spirogyra, Vaucheria, and other algae as
collected, was supplied in such amounts that it was always available to
the tadpoles. A certain number of living and dead small animal bodies
were available as food in this material, and when any of the experi-
mental tadpoles died they were usually eaten by the survivors. For the
most part the tadpole cultures were in pyrex dishes ; but in every case
dishes of the same size and material were furnished to comparable cul-
tures. Most of the cultures were maintained under constant tempera-
ture conditions. These conditions were secured by keeping each dish,
352
EDWARD F. ADOLPH
covered, in a small room which was cooled by air drawn from an adja-
cent refrigerator room by a fan that went into action whenever the at-
mosphere attained a certain temperature. In this way for months at a
time the temperature of the water was kept at 19.0° ± 0.2° C.
The chief observations were made on Rana pipicns ; but supplemen-
tary measurements were carried out on Rana syh'atica, which is char-
acterized by markedly different absolute body sizes.
3WO
3000
2500
2000
1?00
1000
500
0
Age in days
0 20 40 60 80 100 120
Fir,. 1. Growth in weight of Rana pipicns at 19° C. under optimum conditions.
Within the first fifteen days after fertilization the body weight was determined by
weighing 64 individuals together; thereafter single isolated individuals were
weighed. Individual Uj began development on April 9, !''_"'; individual Xa be-
gan on April 25, 1930, and was voluntarily discontinued on June 9, 1930.
NORMAL GROWTH CURVE
Xo curve of growth at constant temperature is known for any
species of amphibian. For this reason the increases of body weight
with time under the optimum conditions of the present experiments are
presented. It mu.st IK- understood that under other conditions, such as
with another food, or in still larger aquaria, or with another race of
Rana pipicns, the rate of growth may be quite different.
The body weight throughout the entire life span is represented in
Fig. 1 for a single individual, from fertilization until after metamorpho-
GROWTH OF TADPOLES AND CROWDING
353
sis. This individual (£//) grew in a volume of 500 cc. of water that
was changed weekly. Before hatching from the egg membranes, tad-
poles increase in weight only very slightly ; this brief period has been
accurately studied by Bialaszewicz (1908). Thereafter body weight
increases rapidly for two or three weeks ; it is believed by Krizenecky
(1917), Faure-Fremiet (1923) and others that during a large part of
this period growth may proceed without food. Thereafter percentage
growth increments decrease markedly. This decrease is probably due
to adverse conditions of unknown kind, since all the cultures were re-
tarded at the same time. But actually no individuals of Rana pipiens
have been cultured without relatively slow growth for some weeks be-
fore the maximum weight was attained preceding metamorphosis. This
°0 2.5 5.0 75 10.0
Volume in cubic mil ii meters
FIG. 2. Frequency distribution of embryo volumes measured on brood E.
Each of the 212 individuals, in the yolk-plug stage, was measured with an ocular
micrometer in two diameters ; the volume of each was calculated as length X square
of breadth X 7r/6.
point will be mentioned again. Metamorphosis is marked by a stop-
page of weight increase and then a sudden loss of over half the body
weight. The loss of weight is one of the first sharp signs of meta-
morphosis that can be detected. The changes at metamorphosis will
be discussed in the next paper.
The present experiments are concerned chiefly with growth up to
1000 milligrams. Figure 1 shows that the rates of growth were not
identical under the optimum conditions in broods U and X.
24
354
EDWARD F. ADOLPH
The first point on the growth curve, which is the egg weight, was
always determined not by measurement of the single egg hut by meas-
urement of a large sample of eggs from the clutch. A frequency dis-
tribution for egg volume is obtained which is always approximately
normal, as is shown for one large sample in Fig. 2. Only small dif-
ferences in mean embryo volume (less than 25 per cent) were found
2000
1000
200
10
'•
Ml
1
J).
1
£
00
1
-a
o
CO
Ace in days
10
30 35 40
FlG. 3. drouth in weight of brood A' at 19° C. under optimum conditions,
is plotted, as in all the subsequent charts, on a logarithmic scale. Solid
points represent 64 or 32 individuals weighed tngi-tlu-r ; open points, single indi-
\ "Itials isolated in 1000 cc. of water.
bet ween tin- first cleavage stages and late gaMrula stages. This conclu-
sion may aK<> !><• drawn from tin- data of l'.ialas/r\vicz (1908) on Rana
ti'in/'oniriu. though some data of Morgan (1906) seem to indicate a
significant innca^c uf total volume during blastocoele formation.
Altogether over 800 embryos of Rana /i/'/1/V;/\ have been measured,
with a mean weight of 5.S" milligrams. Significant differences occur
GROWTH OF TADPOLES AND CROWDING
355
between broods of Rana pipicus from the same pond, the extreme broods
of the ten broods measured having means of 3.30 and 8.35 milligrams.
It was shown by Halban (1910) and Terroine (1921) that egg sizes
are determined in considerable part by the sizes (or ages) of the parents.
Chambers (1908) believed that the size of the individual egg was of
great consequence in the future growth of the individual, but his data
were hardly conculsive on this point.
20
11? -
Apr. 18
Mays
Julq 7
FIG. 4. Growth in weight of Rana pipicns in the wild. Most egg-laying in
the season of 1927 occurred on April 18th in this pond. Group t represents tad-
poles sampled from a circumscribed region of the small stream in question; Group
« represents tadpoles sampled from the whole stream 500 yards in length. Each
weight plotted is the average of 4 to 11 newly collected individuals for t and of
15 to 46 individuals for u. The noon-day temperatures of the water are plotted
also.
A curve for optimal growth (Xa) is plotted upon a logarithmic scale
of body weights in Fig. 3. The most interesting feature of this method
of representation is that in most cultures a certain region of the graph
356
EDWARD F. ADOLPH
of growth is a straight line. " Logarithmic growth " begins at hatching
of the embryo and continues under the best conditions at 19° C. for two
weeks. I hiring this period the body weight doubles every two and a
half days. Thereafter the percentage increment in weight falls off
continuously, though it is sometimes possible to find another straight
line on the logarithmic scale lasting from three until about six weeks
after hatching. The logarithmic scale emphasizes the early parts of
the growth process ; it minimizes to the eye the contrasts in weight that
will be presented below.
o 30 40 60 70 80
Fir,. 5. Growth in weight of brood C at two different temperatures. Each
group or culture contained .^J indi\ iduaN ; a and b were kept at 17° + 1° C., c and
d were kept at 9.5° ± 1° C.
It is -UK-rally considered that ideal growth is logarithmic; that
incrunuit of body substance should every day be proportional to the
substance already present. The maintenance of the logarithmic rate
for th<- two week period at 19° C. is remarkable because of the fact that
during this period the percentage water content of the tadpoles under-
goes huge changes, as is known from the data of Davenport (1897),
Galloway (1900). Sehaper (1902), Bialas/ewic/ (1912) and others.
The d <-d logarithmic rate thereafter might be easily pictured in
terms of Herbert Spencer's (1866) conception that some limiting factor
becomes inadequate to keep up with the mass requirement, perhaps a
GROWTH OF TADPOLES AND CROWDING
factor of intake or of elimination of some substance. It may be said
that such a factor prevents much further enlargement of the tadpole
body, but that after metamorphosis new factors are at work so that
logarithmic growth begins again, judging from the meagre data on the
growth of the adult frog summarized by Donaldson (1911).
Although logarithmic increases of weight are found in many kinds
of organisms, and although logarithmic scales have been used in graph-
ing the present data, it must be stated that there is no intention of
emphasizing those portions of the growth curves in which the logarithm
of the body weight is linear. Some of the data, as those in Fig. 5, may
be accurately represented as parabolic functions. The truth is that the
data are not sufficiently reproducible under diverse conditions of food
and activity, and in diverse broods and species of tadpoles, to insure that
any one formula, or any one controlling factor that it implies, is innately
characteristic of the organism studied. The expenditure of ingenuity
in fitting formulae to the present data is not justified, because of the fact
that growth of an organism is the average of many cycles of mitotic and
incretory activity in the several organs and tissues of the body.
Although no curves for growth in Amphibia were previously worked
out at constant temperature, it is worthwhile to compare Fig. 3 with
those for growth in weight that have been reported. Rana temporaria
(—fusca) as studied by Schaper (1902) showed, in spite of progres-
sively rising temperatures, a progressive falling off in percentage weight
increment from the time of hatching. Bnfo americanns, studied by
Miller (1909), showed rather an increase in the relative increments for
an entire month after hatching. But here it is likely that the tempera-
tures, though not recorded, rose more rapidly. It is apparent also that
the food supply and the aquarium space were more favorable in Miller's
cultures than in Schaper's. Similar results were obtained on Ainbly-
stoina in natural ponds by Dempster (1930). The data on Amblystonia
of Patch (1927) and on Dicinyctyhis of Springer (1909) are insuffi-
cient for comparison.
Some data upon the growth of Rana pipiens in nature were obtained
in the course of the present experiments. Samples of all the tadpoles in
the particular stream from which all the eggs for the laboratory observa-
tions have been collected were taken at weekly intervals during one sea-
son. It should be stated that no eggs, tadpoles, or frogs taken from this
pond were ever identified as being other than Rana pipiens. The curve
shown in Fig. 4 is primarily influenced by the temperature of the stream.
So far as could be observed, the food supply was plentiful throughout
the season, though the population of the pond was large.
EDWARD F. ADOLPH
The effect of temperature upon the rate of growth was studied only
in a preliminary manner in the laboratory. Two cultures (Fig. 5) were
kept in a small undercooled air-bath at 9.5° C. within a refrigerator
room, and were comparable in every respect with two cultures kept in
another air-bath at 17° C., this bath being cooled by a jacket of flowing
water. Throughout the course of growth the former were about half
the weight of the latter, from which it may be concluded that assimila-
tion per unit of mass was half as rapid only in the first week after
hatching. To maintain half the weight later on, however, meant dou-
bling the body weight in the same time interval at both temperatures.
THE EFFECT OF CROWDING ON GROWTH
It was noticed by numerous observers, beginning with Hogg (1854)
and Semper (1873), that individuals of various aquatic species were re-
tarded in growth by confinement in small dishes or by the presence of
many other individuals. It was ordinarily supposed that this influence
was due to fouling of the water by products of metabolism. This ex-
planation did not seem to fit all the facts, though it has recently been
again urged by Crabb (1929); and for Amphibia the suggestion was
made by Pfliiger (1883) that mechanical disturbance was a hindrance
to growth. Yung (1885) experimented with deep and shallow aquaria
of uniform volume, and concluded that surface area of the water was
the primary factor. It may be pointed out that tadpoles kept in deep
vessels are stimulated to greater activity by the shortage of oxygen, and
that the path of locomotion, which is usually horizontal, is shorter. Ba-
bak (1906) noticed that crowded tadpoles were- not only smaller but had
relatively smaller digestive tracts, a conclusion which is not fully con-
firmed by the more extensive measurements of Klven (1928). Bilski
(1921) and Goetsch (1924) tested various devices for overcoming the
possible effect of chemical disturbances.
It seemed that the first step in the analysis of this spatial factor lay
in the accurate measurement of growth rates under conditions of crowd-
ing. For this purpose, body weights were determined at frequent in-
tervals in a number of cultures derived t'nnn a common brood. The
cultures were varied in two ways: (1) in number of individuals per
unit volume of water, and (2) in volume of water per given number of
individuals.
The first mode of comparison is illustrated in Fig. 6. Each culture
here represented was kept at 19° C. in a pyrex di-h with 500 cc. of
water which was changed weekly. The water had a surface of 280
sq. cm. and therefore a depth of 1.8 cm. The retardation in growth is
such that the average weight in the culture containing 64 individuals
GROWTH OF TADPOLES AND CROWDING
359
was, at 19 days after hatching, only one-sixth, and at 28 days after
hatching only one-twelfth the average weight in cultures of one single
individual. Differences of this sort require no elaborate methods for
their demonstration. Four other series of the same sort, grown during
three separate seasons, confirm both qualitatively and quantitatively the
result of Fig. 6. Each series represents two months of care and meas-
urement.
2000
1000
FIG. 6. Influence of density of population upon growtli in weight of brood U
at 19° C. Each culture was in 500 cc. of water, changed weekly, and fed ad
libitum. Numbers indicate the individuals per culture. In the less crowded con-
ditions two duplicate cultures were averaged.
The effect of volumes of water of varying sizes is illustrated in Fig.
7. In these cultures the proportion of surface to volume of water was
nearly constant, so that only the one factor varied. The smallest aqua-
rium was so small that practically all the space was occupied by the food
supply, and the tadpoles were ultimately killed by fouling of the medium.
360
EDWARD F. ADOLPH
Even the increase of the volume of medium from 500 cc. to 1000 cc.
for one tadpole had a marked effect upon body size. The effect of large
volumes (or few tadpoles) became more marked at advanced stages; the
effect in the range of small volumes manifested itself earlier.
The shape of the growth curve was altered by crowding (Fig. 6).
Measurements made early in the life cycle (in brood X) showed that for
the first week after hatching the rates of growth were equal in all cul-
tures. Then, however, the most crowded ones lagged behind and there-
1000
FIG. 7. Influence of size of culture dish upon growth in weight of Rana syl-
"•nlica, brood Q, at 19° C. Each culture contained four individuals and occupied
the volume of water indicated, being changed weekly and fed ad libitum. Two
cultures were averaged throughout.
after proceeded at much slower rates than the isolated ones. Less
erowded ones began some days later to lag, and proceeded thereafter at
intermediate rates. The isolated or single individuals were the last to
depart from the logarithmic rates of growth, and in most instances con-
tinued at rates greater than any others of the brood.
These comparisons may also be made quantitatively. The most direct
method is to pint the mean body weight (//") upon a given day against
GROWTH OF TADPOLES AND CROWDING
361
the number of individuals per culture (»)• It is then found by trial, as
shown in Fig. 8, that the best correlation is between the logarithm of the
body weight and the square root of the number present. In other sym-
bols,
b
log
W = a — b^n-=a-
The volume of medium per individual (T) is the reciprocal of the num-
ber of individuals per volume (;;). The values of the constants a and b
depend on the particular age at which the comparison is made, since they
2000
1000
Number of individuals
0 \ 4 9 16 ~Z3 S£~~ 49 64 81
FIG. 8. Correlation of body weight attained with number of individuals per
culture. The number of individuals is plotted upon a square root scale. The
brood U data are taken from Fig. 6 at 19° C. ; the brood C data were obtained at
inconstant room temperatures, which were, however, alike for all cultures.
are functions both of absolute size and of the relative rates of growth.
Both constants increase with age. A somewhat different formula was
employed by Bilski (1921). If the growth curve could be represented
by a single equation, it would be feasible to have values of a and b to
hold for all ages. But it is obvious that the growth curve itself depends
in fact upon the increasingly severe effect of crowding as the tadpoles
grow older. The inhibition of growth by crowding is manifested only
in proportion as the tadpoles occupy more and more considerable por-
tions of their medium.
362 EDWARD F. ADOLPH
MECHANISM OF THE CROWDING EFFECT
Knowing to what extent diminution of the size of the environment
affects tadpoles at various ages, diverse conditions were tested to find
how the crowding has its influence upon growth. To rule out each
factor in turn required subjection of crowded and uncrowded animals
to experimental devices which are now to be mentioned in more or less
logical order. The conclusions have been previously summarized in
abstract (Aclolph, 1929).
1. Isolation After Crot^dini/. — Individuals that had been in crowded
cultures were isolated singly or in pairs at various ages in culture dishes
of the same size. These individuals immediately assumed rapid rates
of growth, gaining in percentage weight much faster than uncrowded
individuals of the same aye, and almost equalling the earlier perform-
ances of uncrowded individuals of the same sice. Examples may be
seen in Fig. 1 1 , p. 366. Such isolated tadpoles showed the same subse-
quent decreases in growth rate as the uncrowded controls, once the size
at which logarithmic growth usually ceases had been reached.
2. Croi^th in (>!i! Medium. — The water in which many tadpoles had
been crowded was taken, either every day or every week, and given to
>imilarly crowded individuals of the same age and brood. The old
water gave slight inhibition of growth compared to the fresh-water con-
trols in most cases. But the significant result was that the inhibition
was very slight.
In other tests the water from crowded individuals was given to single
individuals of the same brood; one such experiment is shown in Fig. 9.
These again were onlv very slightly retarded in growth compared to
controls in fiv^h water, and the influence was significant only in water
from very crowded cultures when- metabolic substances may have ac-
cumulated considerably.
Individuals that had ^rown in fresh medium but in crowded groups
were isolated in dishes where they were alone but were in old medium.
They began to grow rapidly just as did their controls that were isolated
in fresh medium. These tests proved that nothing added to or sub-
tracted from the water by the presence of other tadpoles was responsible
for the chief inhibitory influence of crowding. It seemed to mean,
further, that no chemical condition whatsoever was responsible for the
iniliifiicc. Nevertheless, further tests were made to rule out other sorts
of chemical conditions.
3. ('ru-^'dint/ -n'ith Oilier Broods. — Groups of two or four individ-
uals of a vouiiLjcr brood or with numerous individuals of another am-
phibian species. The \oiin^er brood of the same species (R. pipiens)
GROWTH OF TADPOLES AND CROWDING
363
had slightly less inhibitory effect than had equal numbers of the same
brood, but there was no doubt of their effect. Other species used to
furnish crowding were Raiia sylvatica and Amblystoma punctatum.
The sylvatica tadpoles were as effective as equal-sized pipicns tadpoles.
The Amblystoma larvae gave inconclusive results because they were not
furnished with sufficient animal food and so did not grow appreciably.
As a result, the experimental sylvatica tadpoles grew almost as rapidly
as their uncrowded controls.
20
10
0 5 10 15 20 25 30 35
FIG. 9. Comparison of growth in fresh water and in water previously inhab-
ited by tadpoles. Cultures s and t each contained 32 individuals ; every four days
their water (500 cc.) was given to g and h respectively, each containing one indi-
vidual. Culture k contained one individual which received 500 cc. of fresh water
every four days. All belonged to brood X at 19° C.
In one series of tests a number of snails or of leeches were placed
with the tadpoles. These tadpoles grew just as rapidly as their controls
of the same brood.
In another experiment an uncrowded individual that had grown
large was placed in the same culture with 32 crowded individuals of the
identical brood. Growth was checked in the large individual. The
364
EDWARD F. ADOLPH
small crmvded ones were already growing so slowly that the additional
crowding was not significant.
4. Confinement. — It has already been noted that reducing the size of
the culture-dish had the same effect as increasing the number of indi-
viduals in the dish. Thus, even a single individual was greatly retarded
if his environment was small. Experiments were now done in which the
volume of water remained large but the animal was confined to a small
portion of it. This was done by suspending a cheesecloth bag in one
liter of water so that the animal could move through only about 300 cc.
of the water. Growth was markedly retarded as compared with the
growth of animals having a whole liter of water in which to move, as
Fig. 10 shows. Animals were later exchanged between bag and no-bag
2000
•
Fi<;. 10. Influence of reducing the free space in a lar-r volume of water upon
growth at 19° C. Each of the four cultures contained one individual of brood X
in 1000 cc. of water; but in c and d a cheesecloth bag confined the individual to
about 300 cc. of that water. On the 28th day individuals l> and r were interchanged
so that /> \\;i> now confined. Thereupon b was retarded in growth while c forged
ahead.
containers. As ran be si-en in the weight chart, the previously retarded
OIK- now .urew faster than its control in another bag; while the newly
retarded one lost weight for a time, subsequently gaining only as fast
as its confined control.
GROWTH OF TADPOLES AND CROWDING 365
Again the conclusion is indicated that not the volume of medium but
the volume through which the animal can move unimpeded is the effec-
tive factor.
5. Aeration. — An obvious possibility, upon the hypothesis that
crowding was a chemical influence, was that lowered oxygen tension or
increased carbon dioxide tension prevailed in the crowded cultures.
No effect of surface area of the water could be found within the
comparatively narrow limits tested. Single individuals in one liter of
water grew at the same rate when that water had a surface area of 133
sq. cm., as when it had an area of 530 sq. cm. and was therefore only
one- fourth as deep.
In another series of cultures the water was continuously aerated by
a stream of bubbles from the compressed-air supply. In them the con-
trast was as great as usual between crowded and uncrowded individuals.
Such tests seem to rule out any volatile substances as essential to the
crowding effect.
6. Addition of Various Substances. — In a few tests the tadpoles
were cultured in media having various materials added to the tap water.
Water in which frogs had previously been for a day or more invariably
killed the tadpoles. This fact contrasts with the harmlessness of tad-
pole excreta ; the acidity of frog urine may have been the important fac-
tor. Small concentrations of urea and of sodium iodide were tested,
but the experiments were not carried on long enough to demonstrate any
effects upon the growth with and without crowding. No differences
could be observed in similar cultures whether in pyrex glass or in soft
glass, nor when extra glass was immersed in the water.
7. Concentration of Food. — It seemed possible that the higher con-
centrations of food that were necessary for maintenance in crowded
cultures were deleterious. In certain uncrowded cultures equally large
excesses of food (green algae) were supplied, but no inhibition of growth
was observed. The presence or absence of rarer kinds of food, such as
minute animals, with the green food apparently had no influence. To
certain crowded and uncrowded cultures liver from various sources was
added as a supplement to the vegetable diet without result. The action
of any disintegration products of the food, which were not large in
amount, are ruled out because the food was replaced more rapidly than
it died.
8. Frequency of Renewing Medium. — A number of the same pos-
sible factors are eliminated from consideration by tests in which the tap
water and food were changed more or less frequently. Cultures con-
taining few and many individuals were changed daily, in contrast with
cultures containing the same numbers changed weekly, as was the rule
366
EDWARD F. ADOLPH
in lu-nrly all the tests. In all cases, as shown in Fig. 11. a slight superi-
ority of size was attained by those whose medium was renewed daily.
This result agrees with that of tests in which previously occupied water
was given to the tadpoles, in showing that substances given off by tad-
poles have demonstrable, but only very slight, inhibitory influences upon
growth.
2000
70 80
Fi<;. 11. InfluciK c of the frequency of chanuini: ihe medium upon growth in
\\eii;lit, in lirood I' at 19° C. in 500 cubic centimeters. Cultures s, t, and u con-
tained four individuals each; cultures q and r contained (4 individuals each. Of
tln-M- .\- and r had fresh water daily; and the others only weekly. Upon the 46th
and 51st days respectively, two individuals from each of cultures q and r were
isolated in 500 cc. and so were able to resume rapid i.Towth. The water in ra was
ewed daily, in </(/ weekly.
Medium occupied much longer than OIK- week, especially bv inter-
mediate sj/rd tadpoK-.,. u-iially became obviously foul, and when actually
allowed 1" remain it killed the tadpoles. It might be assumed, though it
is not proven, that any lethal materials would retard growth in concen-
GROWTH OF TADPOLES AND CROWDING
367
trations too weak to kill. It is surprising therefore to find that they
retard exceedingly little as they do.
9. Flowing Water. — The crucial test for many factors of growth
was to culture the tadpoles in running water. Individuals in varying
numbers were placed in cheesecloth sacks which would just fill 1.5-liter
o
FIG. 12. Growth of tadpoles in flowing water at variable temperatures. The
cultures contained the varying numbers of individuals indicated by the numerals.
Upon the 42d and 58th days respectively, 1 and 4 individuals were isolated from
culture e, and upon the 64th day 1 individual was isolated from culture d, where-
upon rapid growth was resumed by the isolated ones. All cultures were from
brood U in 1500 cc. of rapidly changing water.
beakers, the open tops of the sacks being supported by wires above the
beakers so that water could not overflow the edges. The sacks were
held spread out by frames made of glass rods. Water was renewed in
each beaker at rates of 300 to 600 cc. per minute, day and night, for
several months.
368 EDWARD F. ADOLPH
Attempts were made to regulate the temperature of this water by
first running it through coils immersed in a regulated water-bath. Suf-
ficient heating capacity and coil capacity were not available at the time,
and therefore it was deemed sufficient for the present purpose to allow
the temperature to vary alike in all the dishes. The mean temperature
was much lower in the day than in the night, both gradually rising from
week to week during the spring season.
As the water flowed from the tap into the beaker it was led under
the surface of the water already present, either inside the cheesecloth
bag or else just outside it. In the former case all the tadpoles eventually
perished because nitrogen from the warming water accumulated in their
tissues ; only in the latter manner could this be avoided. It was proven
by watching the convection of colored substances that the water at all
times mixed through the cheesecloth partitions.
Quite unexpectedly (at the time), the contrast in body sizes became
as large in running as in still water. An experiment is represented in
Fig. 12. Nothing could be more convincing than the comparison side-
by-side of the tadpoles in adjacent beakers after one or two months'
growth.
Animals in running water were compared with those in still water
by allowing the temperature to vary alike in both. Those in still water
occupied dishes that floated in running water. It was found in practice
that the floating dishes were always at slightly higher temperatures than
the beakers containing cheesecloth bags. For this reason the tadpoles
in the still water had an advantage. The essential point is, however,
that still water did not sensibly retard growth, as is shown by the data
of Table I.
One other factor that differed for the cultures in running water and
those at constant temperature was exposure to light. The constant tem-
perature room was dark except during those hours each day when cul-
tures wen- being cared for, at which times they wen1 exposed to dim
artificial light. The tadpoles in running water were exposed to indirect
sunlight throughout every day, and never to artificial light. That light
and dark were not significant is attested by the experiments with still
water under the temperature influence of flowing water, and by nu-
merous experiments in which all comparable cultures were carried on
in stagnant water throughout the season under the conditions of a lab-
oratory room. These cultures showed just as significant effects of
crowding as any did ; they arc illustrated by brood C in Fig. 8.
The tests with flowing water demonstrated conclusively that the
essential influence of crowding is physical rather than chemical. Some-
GROWTH OF TADPOLES AND CROWDING
369
how the tadpole is retarded in proportion to its mean free path of move-
ment.
10. Partitioning of Space. — The next step was to measure the
growth response when individuals were separated into small compart-
ments. Four tadpoles, each one in a small cheesecloth bag, were com-
pared with two of the same brood together in one large cheesecloth bag.
All were subjected to the same running water, with the end results
shown in Table I. The separated individuals grew significantly faster
than those that were together, though each one had on the average less
space to itself.
TABLE I
Comparison of tadpoles grown at the temperature of flowing water under diverse
conditions at 45 days after fertilization.
Culture
Number
of
Individuals
Conditions
Final Mean
Weight
in Milligrams
X2n
1
Exposed from 7th to 45th days
Agitation in 1500 cc., flowing . .
289
X2q
4
Together in 750 cc., not flowing .
697
X2r
4
Together in 750 cc., not flowing
617
X2y
4
Together in 750 cc., not flowing
673
X2s
4
Separated in 1500 cc., flowing
1053
X2t
4
Separated in 1500 cc., flowing
1017
X2u
2
Together in 1500 cc., flowing
814
X3g
1
Exposed from 26th * to 45th days
Agitated in 1500 cc., flowing. . . .
738
X3h
1
Alone in 1500 cc., flowing
1154
X3k
16
Together in 1500 cc., flowing
606
X3i
1
Alone in 100 cc., flowing
575
* The mean weight (X3) at the 26th day was 190 milligrams.
This result may seem somewhat inconsistent with any conception of
a spatial factor. While it is possible that some other influence, such as
temperature or food supply crept in unnoticed, the result appears to be
as well substantiated as any others in flowing water. The result does
not seem to be inconsistent with the conception of locomotor disturb-
ances as the crucial factor in crowding. It is confirmed by experiments
in addition to those of Table I in which single individuals in large bags
grew no faster than single individuals in small bags so long as they were
in flowing water.
11. Agitation. — Any influence of size of the environment as a physi-
cal factor must be exerted through some sensory means. A tadpole
might become aware of the extent of its environment through vision,
touch, or muscle senses. Vision is apparently ineffective, because
25
370 EDWARD F. ADOLPH
crowding is just as influential upon growth in the dark. Probably no
other form of distance reception requires serious consideration in con-
nection with growth. Touch would be expected to be effective either
through the contact of other moving objects or in the course of the indi-
vidual's own movements; muscle senses would be informing chiefly dur-
ing the individual's own movements.
If crowding influences growth through disturbance of the passive
individual, such disturbance could be simulated in other ways. Stirring
of the medium or knocking about of the individual seemed indicated.
Arrangements were made to agitate the tadpoles by putting them in
cheesecloth bags that were held expanded and lifting these bags up and
down in beakers of water. The lifting was done rhythmically by a
' windshield wiper," driven by compressed air at rates of 6 to 12 strokes
per minute.
More than a dozen such cultures were set up from time to time; in
most instances the animals were killed by too great violence in shaking,
by the friction of the cheesecloth, or by crushing from the glass weight
in the bag. Young tadpoles were particularly sensitive to mechanical
i riction. The only cases in which growth could be followed for a sig-
nificant period of time were with tadpoles that were allowed to grow to
200 milligrams before being shaken. In these cases growth was re-
tarded (Table I).
Hence agitation prevents growth. \Yhether this is the factor that
prevents growth in conditions of crowding, or whether this is simply a
new form of violence, can hardly be decided. It will always remain
possible that the proper kind and' amount of agitation will not inhibit
growth. Until such a result is realized experimentally, it is permissible
to regard agitation as the same kind of interference as crowding.
12. Narcosis. — It was possible that sensitivity to touch could be
avoided by anesthesia without interfering wholly with growth. Indi-
viduals in crowds we're treated with chloretone in such a way that they
did not move, except in ran- instances, for three day- at a time. During
these three days their controls, in the same-si/cd crowds, were starved.
Then for three days they were taken out of chloretone and with their
controls were fed. Such periods were alternated for several weeks.
Of course no growth occurred in the non-feeding periods; but in the
intervening three-day periods growth regularly occurred. It was not
possible to carry such cultures long enough to obtain significant results;
apparently the dosage of chloretone required varied with the ages of the
tadpoles, for sooner or later the individuals were killed. Here again a
type of experiment that appears on paper as ideal became in practice
useless.
GROWTH OF TADPOLES AND CROWDING 371
13. IiHjcstion of Food and Crowding. — Little observation is required
to notice that per individual crowded tadpoles do not eat so much as un-
crowded ones. This was remarked by previous observers, and Babak
(1906) and Elven (1928) ascertained that the digestive tracts of
crowded tadpoles tended to be slightly smaller and shorter, relatively as
well as absolutely.
\Yhen green alga are being supplied daily to tadpoles it is noticed
that as soon as body sizes have become obviously smaller in crowded
cultures, the food consumption diminishes. It can be seen that this is
so even if crowded individuals are compared with controls of the same
body size rather than with those of the same age. Decreased food con-
sumption is a tangible intermediate factor between growth and crowd-
ing ; it causes decrease of growth rate and it is caused by crowding.
The mechanism of the crowding effect appears to be, therefore, that
agitation or sensory disturbance decreases food ingestion. Hence the
effect is in the first instance upon the behavior of the tadpoles ; it modi-
fies their responses to food. If crowded tadpoles are further observed,
relatively little disturbance is seen ; there is sufficient time for every in-
dividual to eat plenty of food, the animals are simply idle instead of eat-
ing. They resemble children who do not eat but sit idly at table because
exciting events are going on, if a crude analogy be allowed. It is per-
missible to call crowding a psychological factor in growth, so far as is
now apparent.
COMMENT
The effects of crowding have been studied primarily in the tadpoles
of Rana pipiens. In all chief points the mechanism of crowding has
been confirmed in Rana sylratlca. In the literature are recorded partial
similar results upon Rana teniporarla (Babak, 1906; Kfizenecky and
Podhradsky, 1924) and Rana esculcnta (Yung, 1885; Bilski, 1921),
and upon a great number of other kinds of organisms of which the chief
is the snail Lymnea (see Crabb, 1929). It is impossible to say that
these spatial influences are alike in all species ; for in other species they
have not been fully analysed, and there are some indications (Goetsch,
1924) that they are not alike.
The experimental results on Rana pipiens show that chemical effects
of crowding are insignificantly small. In rapidly running water the full
effects of crowding are demonstrable. They can be simulated by agita-
tion of the tadpoles. Both crowding and agitation were observed to
discourage the assimilation of food. The situation is not that food
energy is used up for motor responses to touch instead of being retained
for growth, but that the food is actually not eaten. There is sufficient
EDWARD F. ADOLPH
time available for eating, but the behavior toward food is modified by
agitation.
It would be possible to look upon the decreased ingestion as a meas-
ure of conservation, for in crowded conditions food supply will under
most natural conditions run short. This behavior has the biological re-
sult that little more food than is required for bare maintenance is eaten.
Whether this is a behavior of foresight on the part of a tadpole, no one
can state.
It may lie pointed out that the demonstration, in Fig. 8, that body
weight is denied the crowded tadpoles in proportion to the square root
of their density (\/»), agrees with what might be expected upon the
conception of disturbance by collisions. The number of random colli-
sions within a unit of time would be proportional to \/». This agree-
ment, however, by no means excludes other views.
The occurrence of growth inhibition in nature under conditions of
limited extent of environment has been reported by many observers.
Almost everyone who watches pond life has seen small undeveloped tad-
poles in small ponds upon the same clay that all tadpoles of the same
species have already metamorphosed and left ponds of greater extent.
Of course, mam- factors differ in these situations, and it is almost cer-
tain that no conclusive experiment will occur outside the laboratory.
In non-aquatic organisms effects of crowding have frequently been
observed; indeed, they have often been discussed in man. The only
species for which effects have been measured under highly controlled
conditions is the fly Drosopliila. In it Pearl (1928) demonstrated ef-
fects of crowding upon longevity and upon fertility. He believed that
chemical influences had been ruled out and that crowding exerted its
effect through the psychological patterns of the flies. It may be added
that Pearl's data on fertility fit the formula presented above for the
relation between body weight and density of crowding, substituting the
number of eggs laid for the weight factor.
Crowded tadpole cultures contain individuals highly diverse in
weight. Ordinary observation of this fact is confirmed by numerous
measurements of weight, which will not be presented because they have
relatively small bearing on the problem of the crowding mechanism.
The data were obtained by weighing the smallest and largest individuals
in cultures of 32, in cultures of -4, and in duplicate single cultures. In-
stances where the- largest individual weighed live times as much as the
smallest were found repeatedly. In an equal number of single cultures
belonging to one brood, the largest never differed from the smallest by
20 per cent of the weight on any one day. The variability is remarkably
small during growth under optimal conditions; crowding, and probably
GROWTH OF TADPOLES AND CROWDING 373
any other limiting condition, increases it. The problem of variability
of size was discussed by Krizenecky and Cetl (1924) in an attempt to
relate inequalities of size to " intensity of assimilation." Whether as-
similation be a physiochemical or a psychological phenomenon, their
correlation remains very indefinite except as they express assimilation
in terms of the concentration of food available to the tadpoles. In the
present experiments it seemed simple to picture the development of
inequalities in terms of variable aggressiveness in feeding. Plenty of
food was available, but only the type of behavior that was influenced
less by crowding would allow the ingestion of much food.
Very probably the smaller sizes of densely crowded tadpoles are
accompanied by disproportions of some organs and tissues. Slight evi-
dence has been cited that such is the case for the intestine. The endoc-
rine organs might be suspected of showing deficiencies or hyper-
trophies. Whether such unusual conditions serve as causal or inter-
mediary factors in the control of body size can only be surmised.
The importance of crowding in any experiment having to do with
growth in tadpoles is evident. No conclusions can be drawn from mass
cultures unless both the number of individuals and the total weight of
the individuals present in each culture are equated daily by discarding
appropriate tadpoles each time a death occurs. It must be emphasized
that crowding is proportional not only to numbers of individuals, but
also to the sizes of the individuals. In this way tadpoles that have
grown large will inhibit one another's growth much more than the tad-
poles whose growth has previously been inhibited.
Concerning the problem of body size, it may be said that the tissues
of animals attain a steady size in the adult not because they cannot
grow further, but because their environment prevents them from grow-
ing. This is attested by the whole body of facts obtained through ob-
servation of regenerating tissues and of explanted tissues. Hence it
is the inhibition of growth which is interesting, for many tissues appear
to have the ability to undergo unlimited logarithmic increase. In tad-
poles and other aquatic species it is recognized that not only the body
fluids, but also the environing media, limit the rate and amount of
growth. This variety of influence has now been quantitatively evalu-
ated. It has been found not to be of a direct physical sort, but is effec-
tive because the tadpole is a reacting organism. It is exhibited ulti-
mately in every individual even under optimal conditions. It may be
concluded that useful growth, like civilization, consists not in the limit-
less expression of inherent powers, but in the careful gradation of
activity to fit circumstances. No other form of response would be
equally liable to attain biological success.
374 EDWARD F. ADOLPH
SUMMARY
1. Growth of the tadpole under optimal conditions is very slow be-
tween fertilization and hatching, proceeds with logarithmic increase of
hulk tor about two weeks, and then declines in rate up to the beginning
of metamorphosis.
2. The crowding of many individuals together causes little change
in the initial rate of logarithmic inavasr, but brings on the decline in
rate much sooner and more severely than in isolated individuals. The
>ame effect results from decrease in the volume of water in which the
tadpoles live.
3. Experimental analysis of the mechanism of the crowding effect
shows that the composition of the water itself has no significant influ-
ence on growth. The full effect of crowding is manifested in rapidly
running wau-r, but not when the individuals are partitioned from one
another. A similar inhibition of growth results from agitation of the
tadpoles.
4. The ingestion of food per individual is much reduced by crowd-
ing. The effect is therefore exerted upon the behavior of the tadpole
toward food. The effect is precisely graded with respect to the density
of crowding, so that it is accurately correlated with the physical size of
the environment. It possibly serves as an example of the inhibitions
through which growth is ordinarily regulated.
BIBLIOGRAPHY
ADOLPH, E. F., 1929. The Quantitative Fffect of Crowding on the Rate of Growth
of Tadpoles. Aunt. Rec., 44: 227.
BARAK, F... 1906. Fxperimentelle Untersuchungen iiber die Variabilitat der Ver-
dauungsrohre. Arcli. I:.nt~^'. Mcch., 21: oil.
IliAi.As/Kwicz, K., 1908. I'.fitrajje zur Kenntnis der Wachstumsvorgange bei Am-
phibienembryonen. Hull. Interim!, .lead. Sci. Cracoi'ic, 67. Sci. Math, et
Nat,, Ser. B", 783.
BIALASZEWICZ, K., 1912. I her das \\-rlialtrn des osmotischen Druckes wahrend
der Kntwicklung der Wirbeltierembryonen. Tdl I und II. Versuche an
Hiihner- und Froschembryonen. Arch. l:u!:c. Mcch.. 34: 4S9.
BII.SKI, }•"., 1921. tvber den Kinfluss des I.rl>riiM-;uiincs auf das Wachstum der
Kaulquappen. Pfliiycr's Arch. yes. /'/; v.v /'<•/. . 188: 254.
CIIAMHF.KS, R., 1908. Einfluss der Ki.^msse und der Temperatur auf das Wachs-
tum und die Grosse des Frosches uinl d< -SM-II Zcllcn. Arch. mikr. Anat.,
72 : 607.
CRAP.!!, !•'.. I)., 1929. Growth of a Pond Snail, I.ymnaca sta^nalis appressa, as in-
diratnl l.\ Increase in Sliell-size. lHnl. Kitll.. 56: 41.
DAVENPORT, C. I'.'., 1897. The Role of Water in Growth. Proc. Dost. Soc. Nat.
V//.Y/.. 28: 73-84.
hi MPSTER, \\ . T., 1930. The Growth of l.ar\a- of Ainliystoina maculatum under
Natural Conditions. Biol. Bull., 58: 1SJ.
DOXAI.I' ON, II. II., 1911. On the Regular Seasonal Changes in the Relative
\\(iL'lit of tin- Central Nervous System of the Leopard Frog. Jour.
Morph., 22: 063.
GROWTH OF TADPOLES AND CROWDING 375
ELVEN, E., 1928. Fiitterungsversuche an Kaulquappen, mil besonderer Beriick-
sichtigung der Babakschen Expcrinicnte iiber die Beeinflussbarkeit der
Darmlange durch verschiedene N aiming. Arch. Entw. Mcch., 113: 61.
FAURE-FREMIET, E., ET J. DRAGOIU, 1923. Le premier cycle de croissance du
tetard de " Rana temporaria." Arch. Int. Physiol., 21: 403.
GALLOWAY, T. W., 1900. Studies on the Cause of the Accelerating Effect of Heat
upon Growth. Am. Nat., 34: 949.
GOETSCH, W., 1924. Lebensranm und Korpergrosse. Biol. Zcntralbl., 44: 529.
HALBAN, J., 1910. Die Grossenzunahme der Eier und Neugeborenen mit dem
fortschreitenden Alter der Mutter. Arch. Enizc. Mcch., 29: 439.
HOGG, J., 1854. Observations on the Development and Growth of the Water-snail
(Limneus stagnalis). Trans. Micr. Soc. London, N. S., 2: 91.
KRIZENECKY, J., 1917. Uber das Verhalten lebender Froscheier und Froschlarven
in destillierten Wasser. Arch. Entiv. Mccli., 42: 604.
KRIZENECKY, J., AND V. CETL, 1924. Uber die Abhangigkeit der Variabilitat der
Korpergrosse von dem Grade der Assimilationsintensitat. Arch. mikr.
Anat. Entw. Mcch.. 100: 164.
KRIZENECKY, J., UND J. PODHRADSKY, 1924. Studien iiber die Funktion der im
Wasser gelosten Nahrsubstanzen im Stoffwechsel der Wassertiere. III.
Mitteilting. Zur nnheren Kenntnis der wachstumssteigernden Wirkung
im Wasser geloster Niihrsubstanzen unter normaler Fiitterung mit ge-
formter Nahrung. Arch. ;jes. Physiol., 204: 25.
MILLER, N., 1909. The American Toad (Bufo lentiginosus americanus, Le Conte).
A Study in Dynamic Biology. Am. Nat., 43: 641.
PATCH, E. M., 1927. Biometric Studies upon Development and Growth in Am-
blystoma punctatum and tigrinum. Proc. Soc. Expcr. Biol. Med,, 25:
218.
PEARL, R., 1928. The Rate of Living. New York.
PFLUGER, E., 1883. Das Ueberwintern der Kaulquappen der Knoblauchkrote.
Arch. ges. Physiol., 31: 134.
MORGAN, T. H., 1906. Experiments with Frog's Eggs. Biol. Bull., 11: 71.
SCHAPER, A., 1902. Beitrage zur Analyse des thierischen Wachsthums. I. Arch.
Entw. Mcch., 14: 307.
SEMPER, C., 1873. Ueber die Wachstums-Bedingungen des Lyninacits stagnalis.
Arb. Zool.-Zoot. hist. IVilrzburg, 1: 137.
SPENCER, H., 1866. The Principles of Biology. 2 vols. New York.
SPRINGER, A., 1909. A Study of Growth in the Salamander, Diemyctylus viri-
desccns. Jour. Expcr. Zool., 6:1.
TERROINE, E. F., ET H. BARTHELEMY, 1921. De 1'existence de rapports bio-
metriques entre la Grenouille rousse (Rana fusca) et ses oeufs a 1'epoque
de la ponte. Compt. rend. Acad. Sci., 173: 740.
YUNG, E., 1885. Influence du nombre des individus contenus dans un meme vase,
et de la forme de ce vase, sur le developpment des larves de Grenouille.
Compt. rend. Acad. Sci., 101: 1018.
BODY SIZE AS A FACTOR IX THE METAMORPHOSIS OF
TADPOLES
EDWARD !•. ADOLPH
( /•><>;;; tlic Pliysii'l<><iical Laboratory. The I'ni'rrsity of Rochester School of
Medicine and Dentistry, Rochester, N. Y.)
INTRODUCTION
The role of body size in the activities of organisms has been studied
only in a comparative way. Its effectiveness can be demonstrated ex-
perimentally, however, within a single species in relation to a number
of physiological functions. One of these functions in Amphibia is
metamorphosis.
When many tadpoles were reared in one aquarium, as is described
in the preceding paper, they were retarded markedly in growth as com-
pared with isolated individuals. When the time came for the isolated
individuals to metamorphose, certain of the partially crowded indi-
viduals were also able to metamorphose. But the body weights of the
latter were much less than those of the isolated individuals. Other tad-
poles crowded more densely did not metamorphose at this time, but were
able to metamorphose at later times. To analyse these relationships,
data were obtained under controlled conditions upon the changes of body
weight and the ages at which the morphological changes of metamorpho-
sis occurred.
In the preceding paper si/e was regarded as a result conditioned
inter alia by crowding. In the present paper size is to be considered as
a condition, the result produced being inter alia metamorphosis.
Two species, Rana sylvatica and Rana pipicus, were used in the ob-
servations, and the ages at metamorphosis were recorded for about 190
individuals that had been reared under known conditions. In selected
instances the body weights of single individuals were measured; in other
instances groups of individuals were followed with respect to body
weight through metamorphosis. That crowding resulted in delay in
tadpole metamorpho.si.s \\a> reported by Yung ( 1885), but no data on
body weight were obtained by him. The present experiments were
recently summarized in abstract (Adolph, 1930).
376
SIZE AND METAMORPHOSIS 377
WEIGHT CHANGES DURING METAMORPHOSIS
The progressive changes of body weight in tadpoles crowded to
varying extents are shown in Fig. 1. Some time after the increase of
body weight of the tadpole has fallen below the initial " logarithmic "
rate, the percentage increment is greatly reduced, and then increase
ceases. Finally body weight is lost rapidly for about two weeks, at the
end of which time metamorphosis is visibly complete.
The tadpole has ceased to grow before most of the morphological
changes of metamorphosis are apparent. The first changes are the bud-
ding of the hindlegs ; no others are observable ordinarily until the tad-
pole has begun to lose weight.
After metamorphosis is complete, the body weight of the frog, if
not fed, is almost constant for many days. There is a slight gradual
loss due to the fact that body tissue is being used as the source of meta-
bolic materials. Three or four weeks (at 19° C.) after the maximum
weight of the tadpole is reached, the final weight of the metamorphosed
frog is attained. On the average 60 per cent of the body weight is lost
in this four-week period.
The forelegs usually burst forth from under the skin when about
one-third of the metamorphic loss of weight has occurred. This event
is the most convenient morphological event to identify, and in the
present study it has been used as the criterion of metamorphosis. But
when thus referred to the changes of body weight it is found that the
appearance of the forelegs varies considerably in time of occurrence.
If a single criterion of metamorphosis is required, the best one would
seem to be the point at which the body weight is halfway between the
maximum weight of the tadpole and the weight of the frog three or four
weeks (or an equivalent age at any other temperature) thereafter; for
the change in body weight represents an average of all the changes that
are occurring in the body.
The tremendous loss of weight during metamorphosis does not
represent a 60 per cent reduction of all chemical constituents of the
body. While it is well known that catabolism of nitrogen and other
substances is increased at this time, the losses through catabolism do
not nearly correspond to the losses of weight. The percentage of solids
is known to increase markedly (Schaper, 1902), and this change alone
accounts for a considerable portion of the loss of body weight.
BODY WEIGHT AND METAMORPHOSIS
At 19° C. the maximum weight attained before metamorphosis in
Rana sylvatica, brood Q, was 1230 mg. and in Rana pipiens, brood U,
was 3425 mg.
378
E. F. ADOLPH
Mar. 28
27 May?
Time in
IK;. 1. Growth in weight of five cultures of /UNM syli'titica, brood Q, at 19°
C. The numbers of individuals contained in the cultures are indicated in circles;
each culture was in 1000 cc. of tap water, having a surface of 550 sq. cm. and a
depth nf 1.x rni., that was changed once a week. The first appearance of hindlcgs
in the cult uri- is indicated by a bracket, and the appearance of forelegs, which was
taken as the simi of metamorphosis, is indicated for eacli individual by an arrow.
The li'"ly weights after metamorphosis were determined for smaller groups of
individuals which were numbered in the order of metamorphosis. The subsequent
history of culture c is indicated in Fig. 4.
SIZE AND MKTAMORPHOSIS
379
When the uncrowcled tadpoles of a brood metamorphosed, the
slightly crowded individuals also metamorphosed, and each of the
events marking this transformation occurred upon almost exactly the
same day for all individuals. But, as Fig. 1 shows, the body weights
that had been attained upon the day when the decrease of weight began
were diverse. With a density of fourteen individuals per liter the mean
weight was only two-thirds of the weight where the density was one
per liter.
During metamorphosis the same relative differences of size were
maintained, so that the resulting frogs were of diverse sizes. There
was, however, a slight tendency for the smaller tadpoles to lose a
larger percentage of their body weights in the transformation process.
Hence the percentage diversity of sizes was somewhat greater among
the complete frogs than among the tadpoles.
The great contrast in the sizes of frogs is illustrated by the fre-
quency curves of maximum weight represented in Fig. 2. The sizes
GOO
Weiqht
FIG. 2. Frequency distribution of the final weights after metamorphosis in all
cultures of Rana sylvatica, brood Q, at 19° C. The numbers indicate the initial
densities of the tadpole populations in individuals per 1000 cc.
are thus shown to depend primarily upon the amount of growth that
was previously allowed by the density of the tadpole population (or
other limiting factor).
In addition to knowing the sizes attained by individuals that meta-
morphosed, it is important to know the sizes of tadpoles that did not
metamorphose. Is there any sharp limit of body weight that deter-
mines whether or not metamorphosis shall occur? The maximum
weights attained by the largest tadpoles not metamorphosing and the
other tadpoles metamorphosing are given for one brood of Rana pipiens
in Table I. So far as data are available, they indicate that body weight
constitutes a decisive quantitative factor in metamorphosis. At 19° C.
the upper limit of size that did not allow metamorphosis within 300 days
after fertilization was 2200 mg. for brood U, Rana pipiens (individuals
ta and z>a), and within 150 days was about 550 mg. for brood Q, Rana
380
E. F. ADOLPH
syhiiticn; in the latter brood the tadpoles that did not eventually meta-
morphose were very few.
AGE AT METAMORPHOSIS
At 19° C. uncrowded individuals of Raiia sylratica, brood Q, ac-
quired forelegs at the age of 54 days. In Rana pipicns, brood U, the
corresponding stage was attained at 117 days. The tadpoles that were
slightly crowded were able to metamorphose at the same time as un-
crowded ones. Hence within certain limits the body size had little in-
TABLE I
Ages and weights during metamorphosis of individuals of Rana pipiens, brood U,
at 19° C.
Designation
Age at
Appearance of
Forelegs
Body Weights
on the 105th
Day
Maximum
Body
Weight
Body Weight
at Appearance
of Forelegs
Final Body
Weight After
Metamorphosis
days
"ig.
mg.
mg.
mg.
ca
117
2735
2735
1400
—
3
la
117
118
3425
2730
3425
2770
2450
2060
—
ua
119
—
1510
1300
qaa
ub
120
148
2880
2890
2750
1860
1360
1110
raa
152
—
2860
—
—
oa
225
—
—
—
—
ta
—
1010
—
—
—
va
—
2000
—
rob
343
—
—
—
—
qab
ma
344
457
1585
1930
—
—
—
fluencc upon the time of onset of metamorphosis. But tadpoles that
were densely crowded did not metamorphose at the same age as un-
crowded ones. This is apparent in the two densest populations of
Fig. 1.
If the frequency of various ages at which metamorphosis occurs is
plotted, as in Fig. 3, the contrast is great. The most densely crowded
individuals not only never metamorphosed at so young an age as nil-
crowded ones, but various individuals metamorphosed at highly diverse
times.
The diversity of ages at which metamorphosis occurred is illustrated
in detail in Fig. 4. Over a period of more than two months trans-
SIZE AND METAMORPHOSIS
381
formations frequently occurred in the particular culture illustrated.
Of course, the transformations cannot be said to have occurred at ran-
dom, for in each case it was usually the largest tadpole that began to
metamorphose next.
12.5
150
25 50 75 100
Ace at metamorphosis In days
FIG. 3. Frequency distribution of the times (ages) of appearance of forelegs
in all cultures of Rana syhatica, brood Q, at 19° C. The numbers indicate the
initial densities of the tadpole populations.
CO
Jul. AUJJ, Sep. Oct
Time in months
FIG. 4. Sequence of body weights in culture Qe. Each few individuals that
metamorphosed were weighed after the forelegs had appeared, the individuals be-
ing numbered consecutively as they metamorphosed. The total numbers of indi-
viduals in the culture are indicated in circles. Two average individuals ea were
isolated into 1000 cc. before any had metamorphosed ; these were able to grow con-
siderably before they transformed.
In the brood shown in Fig. 4, the last individual that had survived
came to metamorphose 237 days after its growth started. Its age was
E. F. ADOLPH
then 440 per cent of the age when the uncrowded individuals of the
Bailie In-odd metamorphosed, which may he referred to as "par." In
/\\ina /i//i/V;;.v the last survivor metamorphosed at the age of 457 days
(Tahle I), which was 390 per cent of par. While metamorphoses are
frequent at ages near par, they become less frequent per unit of the
population exposed to metamorphosis as age increases. This is due not
to the death rate among the retarded tadpoles, but to the fact that the
condition which must he met before metamorphosis can occur, which is
body size itself, becomes slower in rate of attainment.
BODY \Yi.n ;IIT AND AGE
The interaction of the two factors of metamorphosis, namely, size
and age, may now be evaluated. It was found, as shown in Table I,
that individuals that were just on the verge of attaining the size neces-
sary for metamorphosis were still able to metamorphose after a delay
of some weeks or months, even though they made little or no further
gain in weight. The charts of bod}- weight indicate that metamorphosis
to the extent of stopping growth in weight might be said actually to
have occurred at par age, but the morphological changes of metamorpho-
sis did not proceed. Evidently, within certain limits, a deficiency in
body size can be compensated by an increase of age.
The way to compare the roles of the size factor and the age factor
in metamorphosis is to plot the two together. This is done in Fig. 5
for the one brood on which most data are available. Since the known
body weights are more numerous after the completion of metamorphosis
than at the beginning of metamorphosis, the final weight of the frog is
used as the measure of body si/.e. The same sort of curve results,
however, whether maximum weight of the tadpole or weight on the day
that forelegs are acquired be- used in place of final weight of the frog.
The best curve drawn empirically through the points of Fig. 5 is a
rectangular hyperbola. If II' is the body weight in milligrams after the
completion of metamorphosis, A is the age at which the forelegs broke
through in days after fertilization of the egg, and c, d, and c are con-
stants, the relationship (A - - c) (IV -d)--c represents the graph.
The constants d and c represent the asymptotes of the hyperbola. The
conclusion may be drawn that no possible increase of size would allow
metamorphosis to occur before c days of age, and neoteny would last
indefinitely if sufficient body size to result in a frog of weight d were
not attained.
For the brood Q at 19° C., living on the diet of Spiroyyra, Vau-
clicriu, and spinach used, the minimum age c is 51 days, the minimum
body weight d is 100 mg.. and the constant c is 1200 day-milligrams.
SIZE AND METAMORPHOSIS
383
The curve as drawn in Fig. 5 represents these values. Under the con-
ditions in which brood Q was reared, the influence of other factors
upon the initiation of metamorphosis was evidently small. Body
weight and age were the effective factors in conditioning the onset of
metamorphosis.
600
to
e
<T3
400
cO
J
cx
|zoo
£
Q;
~r^ f
1
(W-16
OXA-5C
= 1200
\
°A°
o
V
~-oJL °
1
: — o-
0
o
o
0
0
0 0
ifl
?n 90 HO li
at metamorphosis in daus
FIG. 5. Correlation between the ages at which the forelegs appeared and the
weights of the frogs after metamorphosis was complete, in all individuals of Rana
syfoatica, brood Q, at 19° C. The curve drawn through the points is represented
by the formula for a hyperbola, the dashed lines being the asymptotes. The four
solid points represent the modes for various density groups read off from Figs. 2
and 3.
COMMENT
The factors that have been held to be responsible for initiating meta-
morphosis in Amphibia may be roughly classified as : Age, size, previous
history, food, oxygen supply, temperature, hormonal relations, and
heredity. The roles of age and size have been evaluated above by using
observations in which the other factors were held largely constant. If
under previous history be included rate of growth and crowding, then
it has been shown that these are of importance chiefly because they in-
fluence size.
Types of food have not been varied in these experiments; and it
may be that all the observations reported by others in which the food
was varied really influenced metamorphosis either through size or
through endocrine mechanisms. Starvation was reported by Barfurth
(1887) to initiate metamorphosis in frogs. If his data are analyzed,
384 E. F. ADOLPH
however, it is found that by no objective test of significance were his
starved individuals different from his fed individuals. Powers (1903)
concluded from careful observations that sudden starvation precipitated
metamorphosis in Ainbiystoina. Several attempts were made during
the present experiments to bring on metamorphosis by denying food to
tad] idles that had almost attained the minimum size required for meta-
morphosis. But none metamorphosed without further feeding.
High oxygen tensions were stated by Huxley (1925) to inhibit the
metamorphosis of frogs. Extirpation of the lung rudiments by Helff
(1931 ) had no significant effect upon the time of metamorphosis. The
necessity of rising to the surface for air, the amount of gill surface, and
the contact with air are said by Powers (1903) to be of no consequence
in the metamorphosis of Ainbiystoina.
The temperature was held constant in the present experiments.
Uhlenhuth (1919, 1921) reported that when grown at low temperatures,
certain urodeles not merely took longer to attain metamorphosis, but
grew to a larger size before metamorphosis.
The influence of heredity has never been studied in frogs apart from
environmental factors. That broods differ within the same species is
possibly indicated by the varying reports of size at metamorphosis.
Thus, in Rana pipicns Kunt/. (1924) reported that the maximum size of
the tadpoles was 6.8 grams, while Helff (1926) reported that the maxi-
mum size was 1.4 grams. In both cases, nevertheless, 57 per cent of the
body weight was lost before metamorphosis was complete.
The evaluation of size as a factor in determining the onset of meta-
morphosis does not imply that size is an independent variable. \Yhen
all the factors are quantitatively known, it will probably be found that
most of them are both effects and causes. It may be that one or an-
other chemical or physical condition will appear to set aside the usual
complex of conditions. It is already known that thyroid feeding will
render the size factor unnecessary for metamorphosis ; very small and
young tadpoles thereby attain adult properties. But the rate of an
endocrine activity is coordinate!}' correlated with many other factors,
and it would he almost anomalous if it ultimately proved true that a
single limiting factor ordinarily controlled the onset ot metamorphosis.
For anyone who desires to visualize a causal concatenation of fac-
tors, a schema to which the author does not subscribe, a possible mecha-
ni'-iii bv which size and metamorphosis are correlated may be pictured
as follows. It is well known that metamorphosis is often controlled by
the activity of the thyroid gland. This gland is ordinarily thrown into
sufficient activity to induce metamorphosis only in the presence of an-
terior pituitary tissue from metamorphosing tadpoles (Smith, 1923;
SIZE AND METAMORPHOSIS
Uhlenhuth, 1928; Ingram, 1929). But the anterior pituitary also often
controls the rate of growth, and hence the body size (Smith, 1918;
Allen, 1920). The anterior pituitary must ordinarily attain its ability
to set off metamorphosis through its developmental age, but in addition
cannot actually set off metamorphosis unless it and other organs have
attained a certain size through growth. The failure to attain this mini-
mal size or activity, either absolute or relative, may be due not merely
to insufficient nutriment, but equally well to any other influences re-
tarding growth.
Many important physiological reasons may be postulated as to why a
tadpole much smaller than the usual should not metamorphose. It is
doubtful whether the relative objective importance of any of these rea-
sons could be evaluated. In nature all sizes from a few milligrams to
many grams, and all ages from a few days to several years are found to
be sufficient for metamorphosis in one amphibian species or another.
Almost no physiological block to metamorphosis cannot, it may be sup-
posed, be overcome in the course of evolution. Only the situation as
found in particular species can be described as a fit part of the pattern
of existence.
SUMMARY
1. Tadpoles of Rana sylvatica and R. pipicns if sufficiently retarded
in growth by crowding did not metamorphose at the same ages as un-
crowded ones. They were able to metamorphose at their small sizes
at later times. Those only slightly retarded were able to metamorphose
at the usual time, becoming small frogs.
2. Within certain limits a deficiency of body weight is compensated
by a surplus of age, and a correlation of the two factors has been estab-
lished. Through retardation of growth in size the larval stage can be
greatly prolonged. Body size is therefore a tangible quantitative factor
in the complex of conditions which regulate the onset of metamorphosis.
BIBLIOGRAPHY
ADOLPH, E. F., 1930. Body Size as a Factor in the Metamorphosis of Frogs.
Anat. Rcc., 47: 304.
ALLEN, B. M., 1920. Experiments in the Transplantation of the Hypophysis of
Adult Rana pipiens to Tadpoles. Science, 52: 274.
BARFURTH, D., 1887. Versuche iiber die Venvandlung der Froschlarven. Arch.
f. mikr. Anat., 29: 1.
HELFF, O. M., 1926. Studies on Amphibian Metamorphosis. II. The oxygen
consumption of tadpoles undergoing precocious metamorphosis following
treatment with thyroid and di-iodotyrosine. Jour. Exper. Zool., 45: 09.
HELFF, O. M., 1931. Studies on Amphibian Metamorphosis. VI. The effect of
lung extirpation on life, oxygen consumption, and metamorphosis of Rana
pipiens larvve. Jour. Exper. Zool., 59: 167.
26
386 E. F. ADOLPH
HI . I. S., 1925. Studies on Amphibian Metamorphosis. — II. Proc. Roy.
Soc., Ser. B, 98: 113.
. \V. R.. 1929. Studies of Amphibian Neoteny. II. The interrelation of
thyroid and pituitary in the metamorphosis of neotcnic anurans. Jour.
/:'.r/vr. Zoo!.. 53: 387.
KJXTZ, A.. 1924. Anatomical and Physiological Changes in the Digestive System
during Metamorphosis in Rana pipiens and Amblystoma tigrinum. Jour.
.I/.';-/-//., 38: 581.
POWERS, J. H., 1903. The Causes of Acceleration and Retardation in the Meta-
morphosis of Amblystoma tigrinum: a Preliminary Report. Am. A'a/.,
37: 385.
SCHAPER, A., 1902. Beitriige zur Analyse des thierischen Wachsthums. I. Arch.
Entw. Mcch., 14: 307.
• •in, P. E., 1918. The Growth of Normal and Hypophysectomized Tadpoles as
Influenced by Endocrine Diets. Univ. Cal. Pitbl. Physlol., 5: 11.
SMITH, P. E., AND I. P. SMITH, 1923. The Function of the Lobes of the Hypoph-
ysis as Indicated by Replacement Therapy with Different Portions of the
Ox Gland. Endocrinol., 7: 579.
1'iiLENHUTH. !•'.., 1919. Relation between Thyroid Gland, Metamorphosis, and
Growth. Jour, Gen. I'liysii'l., 1: 473.
I'm.KNiirTii, I-"., 1919. Relation between Metamorphosis and other Develop-
mental Phenomena in Amphibians. Jour. Gen. Pliysinl.. 1: 525.
I 'IILKNIIUTII, E., 1921. The Internal Secretions in Growth and Development of
Amphibians. Am. Nat., 55: 193.
L'lii.KMn in. E., AND S. SCIIWARTZBACH, 1928. Anterior Lobe Substance, the
Thyroid Stimulator. Proc. Soc. /i.r/vr. Biol. Mcd., 26: 149.
Y' :.'., E., 1885. De I'influence des variations dti milieu physico-chimique sur le
developpement des animaux. Arch. sci. j>h\s. nat. (Geneve), Ser. 3, 14:
502.
STUDIES ON THE PHYSIOLOGY OF THE EUGLENOID
FLAGELLATES
III. THE EFFECT OF HYDROGEN ION CONCENTRATION ON THE
GROWTH OF EUGLENA GRACILIS KLEBS l
THEO. L. JAHN
DEPARTMENT OF BIOLOGY, UNIVERSITY COLLEGE, NEW YORK UNIVERSITY
INTRODUCTION
Our knowledge of the effect of hydrogen ion concentration on the
growth of the euglenoid flagellates is extremely scanty. Only a few
organisms have been studied from this viewpoint, and in most cases the
results are at best insufficient evidence for definite conclusions. Prac-
tically all the observations of this character have been limited to
Euglena gracilis, except for those of Linsbauer (1915) and Turner (1917)
on unidentified species and for the comparative studies of Kostir (1921),
Mainx (1924, 1928), and Dusi (1930). The particular problem of the
effect of hydrogen ion concentration on the growth of Euglena gracilis
is one that has received considerable attention for several reasons.
The organism is rather unique in that it possesses a very high resistance
to acid solutions, and the literature on the subject is most confusing
due to its contradictory character and to the fact that the results of
most of the writers were obtained by neither accurate nor comparable
methods. In most cases the actual hydrogen ion concentration was not
determined, in some cases organic acids were used, in other instances
the cultures were not bacteriologically pure, and in no case were quanti-
tative methods employed. Therefore it was believed that an investiga-
tion, in which these factors of unknown importance were controlled,
might prove useful in the development of culture methods and in the
further study of the organism; for this reason the present study was
undertaken.
This investigation was performed under the direction of Professor
R. P. Hall, whom the writer wishes to thank for his advice during the
course of the experiments and for his aid in the preparation of the
manuscript.
1 This paper, together with Parts I and II of this series of studies, was sub-
mitted to the Graduate School of New York University in partial fulfillment of the
requirements 'for the degree of Doctor of Philosophy, April 1, 1931.
387
THEO. L. JAHN
HISTORICAL SURVEY
The unusually high resistance of Euglena gracilis to acid solutions
\vas first recorded by Zumstein (1900), who used citric acid in his
cultures in order to reduce the growth of bacteria. He found that E.
gracilis grew well when 1-2 per cent citric acid was added to the 'earth
infusion ' used as a medium. Likewise, he obtained very good cultures
with .5 per cent peptone to which he had added as much as 4 per cent
citric acid. However, he obtained only poor cultures with .5-1.0 per
cent tartaric acid and no growth at all with .2 per cent oxalic acid.
Ternetz (1912) repeated the experiments of Zumstein and found
that citric acid was non-toxic to E. grucilis in peptone, beef-extract, and
earth infusions, whereas it was quite toxic in synthetic inorganic media.
Furthermore, she was able to detect no difference in toxicity between
lactic, tartaric, malic, and citric acids when present in equimolar con-
centrations.
Pringsheim (1912) performed the same type of experiments but
failed to corroborate the findings of Zumstein. He found that when
.5 per cent citric acid was added to the peptone medium, Euglena graci-
lis grew very poorly, but he was able to obtain very good cultures with
.12 per cent or less citric acid in the same type of medium. Therefore
he concluded that the high acid resistance reported by Zumstein was
erroneous, and citric acid was non -toxic only in very high dilution.
Linsbauer (1915), working with an unidentified species which Mainx
(1928) believed was /{. Klebsii, found that citric acid was certainly toxic
in concentrations as low as .07 per cent. Turner (1917), using an un-
identified species of l''.u»lcnti in bacterized cultures, found that an alka-
line medium was favorable for growth of the organism.
Kostir (1921) made a study of the comparative resistance of seven
euglenoids to various concentrations of citric acid. He found that
Euglena gracilis was far more resistant than the other species used.
The order in decreasing magnitude of the resistance of the species
studied was: E. gracilis, Phacus anacoelus, E. oxyuris, E. ehrenbergii,
E. geniculata (?), E. acus, E. descs.
Tannreuther (1923) found that his most healthy cultures of E.
gracilis were strongly alkaline and that the poorest cultures were either
acid or very slightly alkaline. Since his cultures were not bacteria-free,
however, these results might have been due to factors other than hydro-
gen ion concentration.
The next study of the effect of hydrogen ion concentration on
Englcnu grucilis was that of Mainx (1924, 1928), who used bacteria-free
cultures in a medium composed of inorganic salts and .25 per cent beef-
extract. I Ic found that the organisms grew very well in this medium if
EFFECT OF PH OX GROWTH OF EUGLENA 389
it were neutral. He also obtained good cultures when citric acid was
added to a final concentration of 1/400 normal, and fair cultures in
media containing 1/100 normal citric acid. Furthermore, he obtained
very poor growth in cultures containing 1/500 normal NaOH, and no
growth in cultures containing 1/100 normal NaOH.
Dusi (1930), using bacteria-free cultures of Euglena gracilis in a
medium composed of inorganic salts and beef peptone, performed a
more complete series of experiments. The possible effects of organic
acids were eliminated by using only HC1 and NaOH to bring the
medium to the desired pH value. The medium was prepared at six
different hydrogen ion concentrations, the most acid tubes having a
pH value of 3.5-4.0 and the most alkaline ones a pH value of 8.5-9.0.
He found that the maximum density of the cultures was approximately
the same in media with pH values from 4.5 to 8.5, but that the maxi-
mum density was attained sooner in the alkaline cultures. He accred-
its this to a higher rate of division in these cultures. In a later paper
Dusi (1930a) has reported similar experiments with five other species
of Euglena, namely, E. pisciformis, E. stellata, E. anabaena var. minor,
E. deses, E. Klebsii.
At the time the present study was undertaken the question of the
effect of hydrogen ion concentration on Euglena gracilis was highly con-
troversial. The methods and results of Dusi (1930) seem much more
accurate than those of previous workers, but even his results were at
best qualitative and by no means quantitative. Therefore it was
deemed advisable to perform quantitative experiments in an effort to
determine the relationship existing between hydrogen ion concentra-
tion and the growth rate of Euglena gracilis.
MATERIAL AND METHODS
The bacteria-free strain of Euglena gracilis used in this series of
experiments was obtained from the cultures of the Pflanzenphysiolog-
isches Institut of the German University at Prague through the cour-
tesy of Professor E. G. Pringsheim. Fortunately, Euglena gracilis
was much better adapted for experiments of this type than most of the
other available species, because of its more rapid rate of growth in
bacteria-free cultures under known conditions.
The organisms were cultured in 16 X 150 mm. Pyrex tubes plugged
with cotton. The tubes were maintained at a temperature of 28.30
± .05° C. by partial immersion in a water bath designed to accommo-
date a battery of six 100-watt light globes eighteen inches above the
water level. The culture tubes were inclined on a wire rack at an angle
of 45° in order that the plugs would not block the path of the light.
390 THEO. L. JAHX
The medium adopted for the series of experiments was as follows:
KXOa 50 gram
KII2PO< 50 gram
MgSO4 25 gram
\.iCl .10 gram
I'll .05 gram
Partially hydrolyzed casein 5.00 gram
Distilled water 1000.00 cc.
This medium was formulated and selected because the nature and the
relative proportion of the constituents do not change considerably with
titrution or with autoclaving, such as is the case with media containing
ammonium or bicarbonate compounds, which are unstable in alkaline
solutions, or calcium sulphate and phosphate, which are only slightly
soluble in neutral or alkaline solutions. Furthermore, the medium is
well buffered against changes in hydrogen ion concentration within the
range in which it was used. Euglena gracilis may live in such a medium
at pi I 6.7 for four weeks without producing a pH change definitely
detectable with brom thymol blue. The medium was made up in large
quantities and then subdivided and placed in 500 cc. flasks. The
medium in each flask was brought to the desired pH value by the addi-
tion of normal NaOH or normal HC1. The flasks were then plugged
and autoclaved. Kqual amounts (always 10 cc.) of the medium were
then measured directly into the test tubes by means of a Schelbach
side-arm burette graduated to .1 cc. The tubes were plugged with
cotton and autoclaved and were kept in a cool place until used.
Stock cultures for inoculation were grown in 250 cc. Erlenmeyer
flasks in the above medium at a pH of 7.0. Transfers were made from
rapidly <li\ iding stock cultures of 10 to 14 days of age in which practi-
cal !>• all the organism- \\ere in the flagellated condition. Inoculations
were made by means of sterile 12-inch Mohr measuring pipettes of 1 cc.
capacity. The stock culture was shaken for five minutes before inocu-
lations were begun and was then reshaken before each inoculation.
The usual bacteriological method ot aseptic transfer was used.
Measurements of hydrogen ion concentration were made with a La
Motte comparator. The pi I value was determined after inoculation
for one sample tube of each set, and the pi I values were determined for
all other tubes at the end of the experiment. Readings were, in gen-
eral, .HI in, ite to one-tenth of a pi I unit, and the linal values never
\aried more than this amount from the initial pi 1 value except where
other\\i>c stated (.Series Ilia and IVa).
The ability of the organisms to grow at various hydrogen ion con-
centrations \\;ts measured by comparing the initial concentration of
organisms \\ith the concentration in each tube at the end of a definite
EFFECT OF pH ON GROWTH OF EUGLENA 391
time. The same method described in Part I (Tahn, 1929) of this series
of studies was used for counting the flagellates. In all cases the num-
ber of organisms was counted in at least fifty cubic millimeters of each
sample, and three samples were counted from each tube. In all cases
the concentrations of at least two and usually three tubes were averaged
in order to determine the position of each point on the concentration-
pH curve.
EXPERIMENTAL RESULTS
Four series of experiments were performed, and each series will be
described in detail.
Series I
This series was of a preliminary nature. The medium used was the
same as that described above, with the exception that the partially
hydrolyzed casein used was composed of one sample of Difco Trypto-
phane Broth. The stock solution was brought to pH 2.0 by the addi-
tion of normal HC1, and then each flask was brought to the desired pH
value by the addition of normal NaOH. The pH values of the medium
after autoclaving ranged from 3.6 to 8.9. After inoculation at the
beginning of the experiment the range was only from pH 3.9 to 8.3,
due to the neutralizing effect of the 1 cc. of a rich stock culture in the
same kind of medium at pH 6.7 which was used as an inoculum in each
case. Four tubes were inoculated at each pH value to be tested, and
one tube at each pH value was chosen at random and tested colori-
metrically to determine the initial pH after inoculation. Three extra
tubes at pH 7.0 were inoculated so that they could be used to determine
the initial concentration for the series. The average initial count for
the three tubes was .9 thousand per cc., and this was considered to be
the initial count for every tube of the series.
At the end of five days the concentration in one tube of each pH
value was determined. The concentrations in every case were between
5.7 and 6.4 thousands per cc., and, considering the fact that only one
tube of each set was counted, this variation is within the experimental
error and can not be considered further. It was decided to count the
other tubes at a later time when differences, if present, would be more
pronounced. The second count was made on the twelfth day after
inoculation. The results are shown in Fig. 1. The curve shows two
maxima, one at pH 3.9 and one at pH 6.8, and two minima, one at pH
5.5 and one at the highest pH value used, 8.3.
This bimaximal curve was unexpected, and an explanation was not
immediately evident. However, since a trypsin-like enzyme had been
reported for Euglena gracilis (Mainx, 1928), and since the optimum pH
for the digestion of casein may be between pH 6.0 and pH 7.0, it was
THEO. L. JAHN
presumed that the higher growth rate in this range could be explained
on tin- basis of more available necessary food material derived by more
complete digestion of the casein. However, this point was not proven ,
and the high growth rate at pH 3.9 was yet unexplained. It was
thought possible that acid hydrolysis of the casein decomposition
products might have led to the presence of a higher concentration of
available food material in this more acid range. With this in view, a
series of amino-nitrogen determinations were performed on unused
portions of the medium made up at the same time as that used in the
experiments.
The formol titration method of Sorensen was used to determine
the relative amounts of amino-nitrogen present in the samples. Four
determinations were made for each flask of medium tested, and in all
cases 7.0 ± .3 cc. of N '100 XaOH was necessary to restore the pink color
to the solution. These results, of course, showed no significant differ-
ences in amino-nitrogen content of the media at different pH values.
However, only a slight hydrolysis might have given rise to decomposi-
tion products of very high growth-accelerating power, and a slight
hydrolysis could hardly be detected by the method used.
Series II
This series was started before the final results were obtained
from Series I, and it is in some respects a repetition of the former.
However, the results are quite different. The initial concentration of
the organisms after inoculation was 1.8, and the range was from pH 3.9
to pH 7.9. The final concentrations were determined at the end of ten
days. The results are shown in Fig. 1. It is seen that the maximal
growth occurred in the most acid tubes. The minimum present at pH
5.5 in the previous series has apparently shifted to 6.5, and the more
alkaline minimum of the previous series has failed to make its appear-
ance. It was believed that the minimum present in the acid range in
these two series was due to the lack of some particular decomposition
product present in the more acid and in the neutral and alkaline ranges.
Therefore, it was decided to provide the organisms with a more varied
mixture of CUM-MI decomposition products, to make the initial concen-
trations very low, and to make the final counts before the organisms
became numerous enough to exhaust any one type of food material.
Such experiments are described as Series III and IV.
Series III
The method used in this series was the same as that employed in
the two preceding ones. The medium was composed of the same in-
organic compounds, but the partially hydrolyzed casein consisted of
EFFECT OF pH OX GROWTH OF EUGLENA
393
material from three different samples of Difco Tryptophane Broth, one
of which was lighter in color and much more readily soluble than the
other two, and of two samples of Difco Tryptophane Broth which had
been subjected to peptic and tryptic digestion. One of these had been
digested by pepsin for two days and by trypsin for two days : the other
had been digested by pepsin for two days and by trypsin for four days.
These two mixtures and the three samples of Difco Tryptophane Broth
30
20
10
34 5 6 7 8 9 Pri
FIG. 1. Graph showing the results of Series I and II. The concentration of
organisms in thousands per cc. (C) is plotted against pH. Each point represents the
average of the concentrations of organisms in two or three tubes of the same pH value.
were mixed in approximately equal amounts. The stock solution of
the medium was made up at pH 7.0, and then each sample was titrated
to the desired pH value by the addition of normal HC1 or normal
NaOH.
The initial concentration after inoculation was .1 thousand per cc.,
and the pH range was 2.0 to 9.9. The concentrations of organisms
were determined at the end of nine days. The data obtained are
shown in Fig. 2. It is apparent that the organisms grew more rapidly
394
THEO. L. JAHX
between pH 4.0 and pH 7.5 than in the more alkaline range. The
optimum at pH 6.6 is still explainable as being due to the presence of a
trvpi-in-like enzyme with an optimum at pH 6.7 or thereabouts. It is
also evident that no growth took place between pH 2.0 and pH 3.6,
ami that little growth occurred at pH 9.9.
Series IV
This series was, in all essentials, a repetition of Series III. The
initial concentration was .85, the pH range after inoculation was 2.0
to 9.9, and the concentrations of organisms was determined at the end
of eight days. The results obtained are very similar to those of Series
III, and they are also given in Fig. 2. The optimum at pH 6.6 will bear
SERIES IV
2 4 56789 10 PH
I-'K;. 2. Graph showing the results of Series 1 1 1 and IV. The concentration of
organisms in thousands per cc. (C) is plotted against pll. K.irli point represents the
average of the concentrations of organisms in two or three tubes at the same pi I value.
Curve V was obtained by averaging corresponding values of Series III and IV.
the same interpretation as that given above. Since the pH values of
the tubes of Series IV correspond exactly to those of Series III, corre-
sponding values were averaged, and the results were plotted as Curve V
of Fig. 2. The curve shows a decidedly greater amount of growth be-
tween pll 4.1 and pH 7.5 than in the more alkaline range. The opti-
EFFECT OF PH ON GROWTH OF EUGLENA
395
mum is at pH 6.6 and was probably due to the presence of more avail-
able necessary food material produced by the action of the trypsin-like
enzyme.
Series Ilia and IV<i
In each case the concentrations of the organisms in only two or
three of the four tubes inoculated at each pH value in Series III and IV
were determined in order to obtain the curves shown in Fig. 2. The
other one or two tubes of each set were allowed to remain undis-
turbed and were examined at the end of seven weeks. At this time
some of the organisms were encysted on the sides and bottoms of the
tubes, and accurate counts were almost impossible. However, the
results of macroscopic examination and of pH determinations are
shown in Table I. Practically the same results were obtained in both
TABLE I
Initial pH
Amount of Growth
Encystment
Final pH
2.0
—
— .
2.0
3.0
—
—
3.0
3.6
—
—
3.6
4.1
5.2
5.6
++
slight
slight
moderate
4.2
5.2
5.8
6.6
+ + +
moderate
6.8
7.5
+ + +
moderate
7.2
7.7
8.0
+ + + +
very slight
moderate
7.4
7.6
8.5
-f--[-
moderate
7.8
9.0
-f-
moderate
8.5
9.5
±
none
9.5
Key to the amount of growth:
none.
± very slight.
+ slight.
+ + moderate.
+ + + abundant.
+ + + + very abundant.
series, and the two are summarized in the table. These results will
henceforth be referred to as those of Series Ilia and IVa in order to
distinguish them from the quantitative results obtained in Series III
and IV at the end of 8-10 days.
Tests for a Proteolytic Enzyme
In order to confirm the existence of a proteolytic enzyme which
might account for the optimal amount of growth at pH 6.6, inoculations
were made into gelatin and into litmus milk media. Observations at
396 THEO. L. JAHN
the end of four weeks showed doubtful liquefaction of gelatin and no
appreciable effect on litmus milk. However, at the end of twelve
weeks the gelatin cultures were almost completely liquefied, and con-
siderable peptonization of milk and reduction of litmus were quite evi-
dent. These results confirmed the existence of a proteolytic enzyme as
reported by Mainx (1928).
%
DISCUSSION
At the time the present study was undertaken, the question of the
relation of hydrogen ion concentration to the growth of Euglena gracilis
was a highly controversial one. The results obtained by previous in-
vestigators were in a number of cases directly contradictory. The re-
sults of Zumstein (1900), Ternetz (1912), Pringsheim (1912), Kostir
(1921), Tannreuther (1923), and Mainx (1924, 1928), although indica-
tive of the effect of hydrogen ion concentration on E. gracilis, were com-
plicated by at least one other factor such as the use of organic acids,
inaccurate measurements of hydrogen ion concentration, or lack of
bacteria-free cultures. The results of Dusi (1930) are not invalidated
by such factors, but the observations were qualitative only and, as
such, are not very informative as regards the effect of hydrogen ion
concentration on division rate. The present investigation is an at-
tempt to determine in a quantitative manner the relationship existing
between the rate of multiplication of Euglena gracilis and the hydrogen
ion concentration of the medium.
The curves presented in Fig. 2 may be taken as a measure of the
ability of motile stages of E. gracilis to grow in a medium composed of
certain inorganic salts and casein decomposition products at different
pH values, and the curves of the two series of experiments seem to
check as closely as might be expected. The maximum at pH 6.6 is
probably due to the presence of a tryptic-like enzyme which exerts an
optimum action on casein at pH 6.7. The presence of a proteolytic
enzyme in cultures of E. gracilis has been demonstrated by Mainx
(1928), and its existence is confirmed by the gelatin liquefaction and
milk peptonization experiments of the author. The gradual decrease
in the amount of growth with increasing alkalinity as shown in the
curves from pH 6.6 to pH 9.9 checks very closely in both series and is
cjtiitr the type of decrease that might be expected. The sharp rise from
pi I 3.6 to pH 4.1 might possibly be criticized if the range were not so
great, but inasmuch as the range extended from pi I 2.0 to pH 9.9, it was
iidt pi.K t i< ,il»le to use pH intervals smaller than those presented.
However, such a sharp rise in growth-pH curves has been found in the
i ase "t a< id-resistant bacteria. A similar sharp rise has been demon-
EFFECT OF pH ON GROWTH OF EUGLENA 397
strated for Escherichia coli and Bacterium acrogenes (Cohen and Clark,
1919), and therefore it is not surprising that there should be such a
phenomenon in highly acid-resistant protozoa such as Euglena gracilis.
The differences between the results of Series III and IV at the end
of 8-10 days and at the end of seven weeks (Series Ilia and IVa) show
that although the flagellates multiplied much more rapidly in the acid
and neutral media for a short time after inoculation, the maximal den-
sity of population obtained after seven weeks was in tubes of pH
7.4-7.7. Inasmuch as Series III and Ilia were started at the same time
from the same stock culture with the same initial inoculation and were
maintained under the same conditions, and since the only difference
between them is in the length of time the organisms were allowed to
multiply, the shift in the optimum amount of growth from acid to alka-
line media can not possibly be due to an experimental error. This is al-
so true of Series IV and IVa, and the results of Series III and IV and of
Ilia and IVa check very closely. This shift in the maximal amount of
growth is very definite and very consistent in both pairs of experi-
ments.
The results of the present investigation are not in direct contradic-
tion to any of the results of previous workers. However, the fact that
the division rate of Euglena gracilis is initially higher in acid cultures
and that the maximum amount of growth is attained in the alkaline
cultures is very useful in attempting to explain the contradictory and
apparently valid results of previous investigators. The only disagree-
ment between the present results and those of previous workers is with
the results of Dusi (1930), who found that cultures of approximately
the same density (macroscopic appearance) were obtained from pH 4.5
to pH 8.5 in a medium composed of beef peptone and inorganic salts.
However, this might be due to differences in the time of observation in
the two experiments, or perhaps to differences in the medium used.
The reason for such a shift in the maximal amount of growth with
time is a matter of conjecture. One theory which may be presented is
that there was some unknown limiting factor which inhibited growth in
the acid cultures after the first few weeks. However, the possible
nature of such a factor is totally unknown. Another theory which
might be suggested is that the organisms inoculated into the acid solu-
tions were temporarily stimulated to more rapid growth by the acid
and that this stimulus failed to call forth a response after the first few
divisions. However, the possible existence of such a growth-stimulat-
ing power of acid has not been demonstrated, and may not be disclosed
by future investigation. Another theory is that certain hydrogen ion
concentrations might induce temporary encystment with a concomi-
398 TIIEO. L. JAHN
tan! change in division rate. It has previously been observed that
organisms transferred from a neutral medium to a strongly acid one
may experience what has been termed an "inoculation shock" and may
undergo encystment (Mainx, 1928). However, it seems likely that
encystment would induce a temporary decrease in division rate, and
therefore, this theory does not seem to be a likely explanation of the
present phenomenon. If temporary encystment were accompanied by
a temporary increase in division rate, the above results might be ex-
plained. Since practically nothing is known about the relationship
which probably exists between encystment and hydrogen ion concen-
tration and between encystment and division rate, and since encysted
forms were not seen in appreciable numbers in Series III and IV, the
importance of these factors in determining the above shift in maximal
population can not be stated at this time.
The present results indicate that great care should be taken to deter-
mine the time relationships in experiments whose primary purpose is to
determine the relationship existing between growth and hydrogen ion
concentration. This is necessary in order that the early growth rate-
pH relationships as shown in Series I, III, and IV are not overshadowed
by other factors which become noticeably effective during a later period
in the life of the culture, and which might give rise to later contradic-
tory results such as shown in Series Ilia and IYa. It is not clear which
of the two pairs of experiments represents the truer approximation to
the usual growth rate-pi I relationship existing in Euglena gracilis. The
maximal growth in acid solutions as shown in Series III and IV might
be explained as being due to a temporary growth stimulus exerted by
the acid, and the maximal growth in alkaline solutions in Series Ilia
and IY</ as being due to limiting factors which prevented continued
growth in the acid range. I kmever, since it is somewhat unlikely that
a growth-stimulating power of acid, if such exists, would show such a
strong influence at the end of ten days, it seems more probable that
Series 1 1 1 and IV are truer approximations of the usual growth rate-pH
relationship.
SUMMARY
1. The amount of grou th <>l Ruglena gnu His in cultures of different
I>H values has been measured quantitatively at the end of 8-10 days
and has been estimated macroscopically at the end of seven weeks.
2. It is demonstrated that bacteria-free cultures of Euglena gracilis,
in a solution of casein decomposition products and under conditions
which allow mixotropliic nutrition, show, at the end of 8-10days, a high
grou th rate 1 >et u eeii pi I S.'i .mil pi I 7.5 with a maximum at about pH
EFFECT OF pH ON GROWTH OF EUGLENA 399
6.6, and a uniformly decreasing growth rate with increasing alkalinity
between pH 7.5 and 9.9.
3. It is also demonstrated that at the end of seven weeks the most
growth is found to have occurred in the alkaline range, and that the
maximal density of population is at about pH 7.5.
4. It is shown that the results of previous investigators, heretofore
considered contradictory, may be explained on a basis of the time rela-
tionships involved.
5. The existence of a proteolytic enzyme in cultures of E. gracilis is
confirmed.
LITERATURE CITED
COHEN, B., AND W. M. CLARK, 1919. The Growth of Certain Bacteria in Media of
Different Hydrogen Ion Concentrations. Jour. Bacterial., 4: 409.
Dusi, HISATAKE, 1930. Les limites de la concentration en ions H pour la culture
d'Euglena gracilis, Klebs. Compt. rend. Soc. Biol., 103: 1184.
Dusi, HISATAKE, 1930a. Limites de la concentration en ions H pour la culture de
quelques Euglenes. Compt. rend. Soc. Biol., 104: 734.
JAHN, THEO. L., 1929. Studies on the Physiology of the Euglenoid Flagellates. I.
The relation of the density of population to the growth rate of Euglena.
Biol. Bull., 57: 81.
KOSTIR, W. J., 1921. The Comparative Resistance of Different Species of Euglenidae
to Citric Acid. Ohio Jour. Set., 21: 267.
LINSBAUER, K., 1915. Notiz iiber die Saureempfindlichkeit der Euglenen. Osterr.
bot. Zeitschr., 65: 12.
MAINX, FELIX, 1924. Kultur und Physiologic einiger Euglena- Arten. (Vorl. Mitt.)
Lotos, 72: 239.
MAINX, FELIX, 1928. Beitrage zur Morphologie und Physiologie der Eugleninen. I.
Teil. Morphologische Beobachtungen, Methoden und Erfolge der Rein-
kultur. II. Teil. Untersuchungen iiber die Ernahrungs- und Reiz-
physiologie. Arch. f. Protist., 60: 305.
PRINGSHEIM, E. G., 1912. Kulturversuche mit chlorophyllfiihrenden Mikroorganis-
men. II. Zur physiologic der Euglena gracilis. Beitr. z. Biol. d. Pflansen,
12: 1.
TANNREUTHER, G. W., 1923. Nutrition and Reproduction in Euglena. Arch. f.
Entw. d. Organism., 52: 367.
TERNETZ, CHARLOTTE, 1912. Beitrage zur Morphologie und Physiologie der Euglena
gracilis, Klebs. Jahrb. f. wiss. Bot., 51: 435.
TURNER, C. L., 1917. A Culture Medium for Euglena with Notes on the Behavior of
Euglena. Anal. Rec., 12: 407.
ZUMSTEIN, H., 1900. Zur Morphologie und Physiologie der Euglena gracilis, Klebs.
Jahrb. f. wiss. Bot., 34: 149.
TI 1 K MUSCULAR ACTIVITY AND OXYGEN CONSUMPTION
OF URECHIS CAUPO
VICTOR E. HALL
(Prom the Physiological Laboratory of the Hopkins Marine Station, Pacific
Grove, Calif.)
I. INTRODUCTION: NATURAL HISTORY
Urccliis caiifo, a large marine echiuroid worm recently discovered
on the California coast by Fisher and MacGinitie (1928), presents by
virtue of its habits of life a unique opportunity for the quantitative
study of the interrelations between muscular activity, rate of metabolism
rind the mechanism of exchange with the environment.
The animal digs and lives in a U-shaped burrow in the mud of
shallow estuaries, leaving it only occasionally to construct a new burrow.
The upper ends of the burrow open freely to the water. The requisite
exchanges with the environment: respiratory, nutritive, excretory and
reproductive, are accomplished by the animal forcing a stream of
water through the burrow. The movement of water is produced by
peristaltic waves in the musculature of the body wall, originating at or
near the anterior end, and passing posteriorly. The integument in the
region between two consecutive waves is pressed closely against the
wall of the tube. Accordingly, water between the integument of the
constricted regions and the sides of the burrow is carried posteriorly
with the peristaltic wave. The worm from time to time turns around
in the tube, thus reversing the direction of the stream.
The mode of feeding is unusual. Near the anterior end of the
worm there is a ring of specialized mucous glands. The animal presses
the body wall in the region of these glands firmly against the side of
the burrow; then, as the glands secrete, it hacks away, leaving a tube of
mucus attached to the burrow at one end, and to its integument at the
other. The peristaltic movements, usually suspended during the forma-
tion of the tube, are no\v resumed, drawing a stream of water through
the mucus tube, which acts as a filter. Particles over one micron in
diameter an- retained. After filtration of water has continued for some
time, the worm moves forward, seizes the tube with the proboscis and
s\vallo\\-s it whole. Since the food consists of particles included in
the detritus of the estuary bottom, this mechanism enables the animal
400
ACTIVITY AND O, CONSUMPTION OF URECHIS 401
to obtain nutriment without leaving the burrow. The above descrip-
tion of the animal's habits is adapted from Fisher and MacGinitie's
account.
UrccJiis may be kept in the laboratory indefinitely if placed in glass
U-tubes of dimensions approximating those of the burrow and if given
access to aerated sea water. Several specimens, introduced into such
tubes over three years ago by Professor MacGinitie of the Hopkins
Marine Station, are now in excellent condition. Their behavior in the
laboratory is consistent, as far as is known, with that in their natural
habitat.
II. ACTIVITY: VOLUME OF WATER PUMPED THROUGH TUBE
The volume of water pumped through the tube in which the animal
is living is of interest from two viewpoints : First, since the peristaltic
activity of the body wall musculature constitutes by far the greatest
part of the muscular activity of the animal, a measure of the volume
pumped may be regarded as an approximate indirect measure of the
total muscular work. Second, since all exchanges with the environ-
ment are mediated through this stream, its measurement yields data
relative to the potentially available oxygen and food supplies, and to
the facilities for disposal of metabolites and reproductive products.
The method of measurement of the volume pumped is closely
related to that devised by Galtsoff (1928) for the study of the flow of
water produced by the gills of the oyster.
The apparatus is diagrammed in Fig. 1. One UrecJiis (A) was
introduced into a glass U-tube (B), about 2.5 cm. in diameter, the
length of the horizontal segment being 30 cm. and that of each vertical
segment 25 centimeters. The tube was placed in an aquarium of
approximately 100 liters capacity, through the glass front of which
the animal could readily be observed. A stream (/) of aerated sea
water, filtered free of food materials, entered the aquarium continu-
ously and overflowed from a fixed aperture (C), thus maintaining a
constant level in the aquarium. The temperature in the aquarium
ranged from 15.4 to 18.8° C., the average being 16.9 degrees.
The ends of the U-tube projected above the level of the water in
the aquarium. A siphon tube (D} admitted water from the aquarium
into the artificial burrow at one end. The water, forced through the
tube by the work of the worm, passed by means of a second siphon
at the other end into an Erlenmeyer suction flask (£). The flask was
so adjusted that, when filled until water overflowed through the side
tube (F), the level in the flask was the same as that in the aquarium.
Since the level of water in all vessels was the same, the only factor
causing water to move was the pumping activity of the worm. The
27
402
VICTOR E. HALL
overflowing water was caught in a graduated cylinder (G). Collections
were made, as a rule, for five-minute periods. The rate of flow was
ressed in cubic centimeters per minute.
Since the animals frequently turn around in the tubes, it was neces-
sary t<> duplicate inlet siphon, outlet siphon and overflow flask, so that
t IK- flow could be measured in either direction. The siphons not neces-
sary at the moment were closed with pinchcocks. For simplicity, tin-re
is represented in Fig. 1 only that portion of the apparatus required for
measurement of the flow in a single direction.
D
Tr
•
.
I-'n;. 1. Apparatus for measurcim-nt of volume of water pumped by Urechis.
ription in text.
The high degree of variability in the rate of pumping which was
found made it seem wise to make a number of observations on a few
individuals over a considerable period of time rather than a few obser-
vations on each of a large number. Accordingly, only two animals
were employed, both near the average size of mature worms, i.e., about
60 grams weight. Worm I was perhaps twenty per cent larger than
Worm II.1
1 PmiYssor G. E. MacGinitie of the Hopkins Marine Station, Stanford Uni-
versity, kindly provided me with the following weights of ten mature specimens
nf Urechis:
Avci
grams
Maximum,
grams
Minimum,
grams
Total body weight.
62.5
82 4
35.1
\\Vinht without blood
40.7
53 1
21.1
Weight of blood
21.8
31.3
12.5
ACTIVITY AND O= CONSUMPTION OF URECHIS 403
The worms were kept in their tubes undisturbed throughout the
whole duration of the experiments, — about two months. A period of
about a week was permitted for adaptation to their new environment
before collection of data was begun.
Results
\Yith the animals under constant standard conditions, the rate of
pumping during five-minute periods ranged from 0 to 50 cc. per minute.
Two factors in the production of this variability were noted: (1) a
consistent increase during the feeding periods — a factor which has been
studied in some detail; and (2) long periods of inactivity during which
the worm lies in a cylindrical form with integument in contact with the
glass over its whole length and shows no movement. Such periods may
last from twenty minutes to well over an hour. They are usually
terminated by the worm turning around, arid then resuming pumping.
Concerning the significance of these periods of inactivity no suggestions
are offered.
However, even if these sources of variability be excluded by the
choice of non-feeding periods during which the worm was continually
active, there remains a high unexplained variability. For example, in
the case of Worm I in twelve consecutive five-minute periods, in all of
which it was active and during which no feeding occurred, the rate of
pumping ranged from 8.3 to 29.2 cc. per minute, the mean being 19.3
and the average deviation from the mean, 5.0 cc. per minute.
Average Volumes Pumped
In Table I the results of a number of experiments are tabulated.
The first three columns contain the data for the whole period of each
observation ; the second three columns, the data for the portion of the
period during which the worm was feeding ; and the third three columns,
the data for the portion during which it was not feeding. The last
column shows the ratio of the rate of pumping when the worm was
feeding to that when it was not feeding. The averages are weighted.
It will be seen that the larger worm averaged throughout the experi-
ments 16.5 cc. per minute; the smaller, 10.2 cc. per minute. The
variability among the averages of the experiments is considerable.
404
VICTOR E. HALL
pa
"8
•si
<o
<u
I
•&
<u
5
M
c
b
.s
o c
P
*|
T3
1
O
OO too *-" c^j oc <"^i **
oo t^-oo oo *-«,-( ,~i ON
' i— i ^- ^ ^-t ojojtO'-H
>o
<*5
tN-
Z
O
J*
tN
*
"7---" - --"*-
c
•3
1
.§
•jccc^oc^--— ' r^r^rviio
c
H
fc^-^^nrf^H T-IT^^H
o
(*O ^O ^^ ^^ t^1* ^3 ^^"> t*"* f"O
*Q
u O^ ^ r^i r^- ^H o* oo ^t< oo
^
v- CN) ON! lO tO OC O O *O CO
CNCNrfCN'— t ^-iT-t^H^H
sj
O
I
1 ^— » ^ lO O to ro CN ro
^ co <r;rc ro CNCNi-iCN
tN
tN
M
c
0)
•5
a
.510 T^r*— oOc^ir^^H
1 *~^ t^ CO to ^*1 fO 4^^
^^* ^O i^^~ ^^ ^^ I**** ^^
"o
^Ir-j 10— O-nO«O
^ 1 ^O OO re ^ LO Tt* ^O
^N »— 1 ~H »- C
0)
w
•S>OlOroO"^ >O OCvOro
^. vc oc o •* -r o t-~' tN i* \d
O
0
0>
-'OOOvOOlOC^ O'f'^f^O
•5 ro rl< ro OO CS lOtN-^oo
H
H
tn
OD
0
0)
(t
n!
V
u
(!)
_;
bo
rt
5
v- tN r^ w/ *^* ^^ QJ *o >o ^D f**
i_
0)
rs c*l *^f ^* ^O ^ '~v' ^^ tN *^5
>
<
<
^^n^tn -Htv, ro^
*
1
.
I
I
( t , ,
c
1
(
Z E
(I
j
S 0
bs ^
ACTIVITY AND O, CONSUMPTION OF URECHIS
405
Feeding Cycles
The following descriptive data were obtained from a series of
twenty-five cycles during which the worms were observed to be feeding.
Frequency of occurrence: Worm I produced a tube on the average
0.7 times per hour; Worm II, 1.5 times per hour.
Duration of feeding periods, from completion of tube to swallowing
of tube: averages, Worm I 18.3 minutes; Worm II 8.7 minutes. It is
interesting to note that the number of minutes spent in feeding per
hour of elapsed time is closely similar in the two animals: Worm I 12.8
minutes per hour; Worm II 13.1 minutes per hour. The larger worm,
which fed much less frequently, compensated by greater duration of
each feeding period.
Course of activity during the feeding period: Onset: In 18 of the
25 cycles examined there was a decrease in the volume pumped during
the five-minute period during which the tube was formed. The actual
SO
40
l_
<L>
^ 30
6
TD
<3J
I 2°
a
c
3 10
I
0
10
20
30
40
50
60
70
80
90
Time, minutes
FIG. 2. Rate of pumping by Worm I for a 95-minute period, during which
two feeding cycles occurred. The rectangles A and B indicate the time between the
formation and swallowing of the tube in each cycle.
formation of the tube occupied about 30 seconds — during which time
active pumping was suspended. Course: In the period during which
the tube is present, there was observed in 23 of 25 cycles a clear-cut
increase in the rate of pumping as compared with the rate before and
after the feeding period. There is often a step-wise increase during
the first two or three five-minute periods to an irregular plateau in the
curve of pumping rate. End: Immediately or within five minutes after
the swallowing of the tube there is commonly, but by no means in-
406 VICTOR E. HALL
variably, a decrease in the rate of pumping to below 5 cc. per minute.
In Fig. 2 there are represented graphically two consecutive cycles which
occurred relatively closely together.
.li'craijc of activity during the cycle as compared ivitJi that during
inter:\ils between cycles: These data are included in Table I. The
marked increase during feeding is clearly evidenced by the fact that
the average ratio of the rate of pumping during feeding periods to that
during non-feeding periods is, in AYnrin I 1.8; in Worm II 2.3.
('•unparable increases are present in all experiments without exception.
In only two of the 25 cycles examined was there no increase. In one
of these periods the worm ate the tube within half a minute of making
it. In the other, an increase in activity occurred before the feeding wa-
observed. Here it is possible that the tube had been formed earlier
and was overlooked for some minutes.
Discussion
Since the current of water pumped through the burrow finds its
significance to the organism by making possible exchanges of materials
with the external environment, a discussion of its role with respect to
certain of such materials is pertinent.
(1) O.vycjcn. The respiratory significance of the current has been
discussed by Red field and Florkin (1931 ), their discussion being based
in part upon the results communicated in this paper. These authors
point out that the animal utilizes only one-third of the oxygen in the
water inhaled into the hind-gut. Accordingly, at the normal rate of
oxygen consumption 0.013 cc. per minute, a hind-gut ventilation of
6.9 cc. of water would be necessary. Since the average rate of pumping
amounts to about thirteen cubic centimeters per minute, the current is
about twice that necessary for the maintenance of normal respiratory
relations.
In an attempt to determine the mechanism of adaptation of the
worm to waters of low oxygen content, in eight experiments conducted
with the apparatus described above, the worms were given access for
approximately an hour to sea water boiled until its oxygen content was
reduced from about 4.6 to about 2.5 cc. per liter. The oxygen pressure
was thus reduced to about seventy millimeters 1 Ig. The pi I of the
sea water, increased by the boiling, was readjusted t<> the normal value
of 8.2 by addition of a small quantity of dilute' hydrochloric acid. The
activity of the worms under these condition-; was compared with that
during similar hour periods, immediately before and after, during which
normal sea water entered the tube. No consistent effect was observed,
ACTIVITY AND O2 CONSUMPTION OF URECHIS 407
activity being greater in four experiments, unaltered in one and de-
creased in three. The average, however, is 40 per cent greater than
that of the control periods.
As will be described in Part III of this paper, a reduction of oxygen
pressure to 70 mm. Hg is accompanied by a reduction of oxygen con-
sumption to about fifty-five per cent of that in normal sea water.
The fall in oxygen consumption is of itself adequate to compensate
for the decreased amount of oxygen available in the water. This fact,
rather than a consistent increase in the current of water pumped
through the burrow, appears to be the adaptive response of the animal
to water of low oxygen content. For further discussion of this matter,
see Redfield and Florkin (1931).
(2} Food. The food requirement of the animal (expressed in
some such units as calories per hour), together with the food value of
the sea water (in calories per liter), determine the volume of water
(liters) which would be required to be filtered in order to meet the
requirement. It is conceivable that this might be accomplished by
means of a continuous stream of constant intensity. However, Urechis
instead employs the same device as the higher animals, that of periods
of intense food-getting activity alternating with periods in which the
animal is freed for other activities. Thus, Urechis spends only about
one-fifth of its time in the obtaining of food. In order to accomplish
the required filtration within this restricted time, a relatively high degree
of activity is necessary. Unfortunately the data are not available
which would make possible an assessment of the significance of the
magnitude of the stream for feeding as has been done for oxygen.
Under the conditions of the experiments the worms were provided
with sea water so filtered as to be practically devoid of food value.
They had been, and were, accordingly, in a state of chronic starvation.
Whether this would serve to evoke a maximum intensity of food-getting
activities, or would rather, after a time, cause decreased activity and
reduced rate of metabolism as occurs in the chronic inanition of mam-
mals (Lusk, 1928), is not known. However, the fact should be borne
in mind in any attempt to apply the data to Urechis in its normal habitat.
The stimulus provoking the feeding reaction is not known. That
it is not of external origin is shown by the fact that the two worms,
in similar tubes side by side in the aquarium, subjected to the same
environmental influences, including light, jarring, etc., and receiving
the same sea water, carried out their feeding reactions totally inde-
pendently of each other in time. The stimulus is probably of internal
origin.
If, during the period of feeding, a relatively minor mechanical
408 VICTOR E. HALL
disturbance be brought about, such as gently moving the inlet siphon
tube, the worm abruptly stops pumping, casts loose the mucus tube and
back- down into the horizontal part of the U-tube.
III. OXYGEN CONSUMPTION
The oxygen consumption of L'rccliis was determined by two methods
which yielded similar results: (1) a worm, active in a U-tube under the
conditions described above, pumped water from the aquarium into the
Erleinneyer suction flask which had previously been filled with mineral
oil, so displacing the oil with water. The oxygen content of the water
in the U-tube in front of the worm (incoming water) and of that in
the flask at the end of the period (outgoing water) was determined by
the method of \Yinkler (1888), samples being withdrawn by means of
siphons (H and / of Fig. 1) without disturbing the animal in any way.
Knowing the oxygen content of the incoming and outgoing waters and
the volume pumped, the oxygen consumption could be readily calcu-
lated.
(2) A worm was placed in a jar containing approximately three
liters of sea water, over the surface of which a layer of mineral oil
about one-fourth inch thick was floated. Samples of water were with-
drawn at intervals by means of a siphon and their oxygen content
determined by the Winkler method.
Results: First Method
In Table II are tabulated the oxygen consumption (in cc. per min-
ute), the oxygen partial pressure in the incoming water (mm. Hg),
the oxygen content of the incoming and of the outgoing waters (cc.
per liter), and activity or volume of water pumped (cc. per minute).
The experiments are arranged in order of increasing activity. It will
be noted that there is a general tendency for the oxygen consumption
I" increase with increasing activity, as would be expected. There is,
in these experiments, no consistent relation between the oxygen partial
pressure of the incoming water and the oxygen consumption of the
animals. The oxygen consumption rate of the two worms is almost
identical, being 0.0130 and 0.0136 cc. per minute respectively.
Second Metliod
The oxygen consumption of the animals kept in jars under oil was,
during the initial period of each experiment, as follows: 0.0141, 0.0173,
0.02X1 and 0.0120; average, 0.0179 cc. per minute. These values are
definitely higher than those obtained for the oxygen consumption of
ACTIVITY AND O, CONSUMPTION OF URECHIS
409
worms in the U-tubes. The difference is attributable to the exaggerated
peristaltic activity exhibited by the animals in the jars. Apparently,
the absence of the normal contact of the integument serves to activate
the animal's movements. These values are accordingly considered less
representative of the metabolic rate under normal conditions than those
obtained by the U-tube method.
A comparison of the oxygen consumption of Urcchis with that of
closely related forms determined by other workers follows.
TABLE II
Oxygen Consumption of Urechis in U-tubes
Animal
Oxygen
Consumption
Incoming water
Outgoing
water
Activity
Oxygen
Partial
Pressure
Oxygen
Content
Oxygen
Content
cc.lmin.
mm./Hg
cc. /liter
cc. /liter
cc./min.
I.
0.0091
138.2
4.90
2.80
4.3
0.0085
127.2
4.51
3.51
8.5
0.0154
130.0
4.61
3.17
10.7
0.0123
191.8
6.80
6.10
19.5
0.0150
93.3
3.31
3.01
50.0
Average
0.0178
94.2.
133.7
3.34
4.74
3.02
4.18
54.3
17.9
0.0130
0.0097
II.
0.0118
96.0
3.40
2.92
24.7
Average
0.0194
96.2
3.41
2.92
45.2
0.0136
Comparison of Metabolic Rates of Certain Invertebrates
Animal
Lumbricus
Glycera siphonostoma
Hirudo
Sipunculus nudus
Urechis
Author Oxygen Consumption
cc. Oz gm./min.
Averaged results of Thunberg, Lesser and
Konopacki, quoted by Krogh (1916) 0.00189
Cohnheim (1911-12) 0.00123
Montuori (1913) 0.00025
Rogers (1927) 0.00052
Cohnheim (1911-12) 0.00082
Present author . 0.00021
Urechis thus possesses a metabolic rate of magnitude comparable
to related forms but distinctly lower. This is in part attributable to
the fact that this animal has a blood volume disproportionately great
for its size as compared with allied forms. Thus, an average-sized
410
VICTOR E. HALL
animal weighing 62.5 grams possesses blood weighing 21.8 grams.
Although the corpuscles are true cells, it is doubtful whether their
metabolism would give to the blood a rate of oxygen consumption per
-ram comparable to that of the fixed tissues. If the total oxygen
consumption be calculated on a basis of fixed tissue weight, it becomes
0.00033 cc. per gram per minute.
280
6 24°
^—
X
e
: 200
o
a
160
.0
— -i
a
|
I 120
C
o
| 80
40
n
/p
/
/
f
/•° /
/7
/
/
/
/
/ /
7"
x
/*
/
/-
/ /
/
/
x
//
'-j^-~
s
---'""*
—""^
20
40
60
80
100
120
Parii.il Pressure of Oxygen — mm. Hg.
IK;. 3. Relation of oxygen consumption to oxygen partial pressure. Urcchis.
The letti is at the extremities of the curves indicate the correspondence of the
curves to tin experiments reported in Table III.
Consumption and 0.\'\<if>i Pressure
In the experiments in which worms were placed in jars containing
approximately three liters of sea water, the oxygen content of the water
was determined at intervals as it fell due to the metabolism of the
animals. In several cases the rate of fall was accelerated by placing
three worms in a jar instead of a single one. When the oxygen content
had fallen to about 0.2 cc. per liter, the experiment was discontinued.
ACTIVITY AND O, CONSUMPTION OF URECHIS
411
The worms were still active, as shown by persistence of spontaneous
peristaltic activity.
To determine the oxygen partial pressure of the sea water, the
oxygen content was plotted against time, and the oxygen content read
off the curve for the middle of each period between successive samplings.
Under the conditions of the experiments, namely, at a temperature of
17° C., an atmospheric pressure of 760 mm. Hg, and a sea water
TABLE III
Oxygen Consumption and Oxygen Pressure
Experiment and
period number
Duration
of period
Initial Oj
content
O: pressure
at mid-period
Oxygen
consumption
hrs. min.
cc./l.
mm.Hg
cc./min./worm
A: 1
21 30
5.15
97.6
0.0118
2
44 00
1.78
29.0
0.0024
3
4
B: 1
6 30
0.28
0.12
5.19
5.6
0.0013
24 00
79.0
0.0039
2
4 10
0.42
8.5
0.0011
3
4
C*: 1
22 00
2 15
0.18
0.10
3.60
3.9
0.0001
66.0
0.0173
2
1 00
1.08
24.5
0.0063
3
1 05
0.66
14.4
0.0037
4
1 00
0.37
9.9
0.0006
5
6
D: 1
1 05
0.33
0.28
5.24
8.5
0.0006
1 15
116.8
0.0281
2
2 05
3.04
62.6
0.0120
3
1 00
1.40
28.2
0.0115
4
5
0 55
0.61
0.42
14.4
0.0028
* At the conclusion of Experiment B, the greater part of the water was removed
from the jar and fresh water substituted. The earlier parts of Experiment C then
represent post-anoxybiotic metabolism, the more active oxygen consumption sug-
gesting that an "oxygen debt" was being made up.
In Experiment A one worm was used, in the remaining experiments, three.
chloride content of 19 grams per liter, the oxygen content of water in
equilibrium with atmospheric air is 5.66 cc. per liter (measured at
N. T. P.). (Fox, 1907.) Under these conditions the oxygen partial
pressure is 159.6 mm. Hg. The partial pressure of any sample of such
water of which the oxygen content is known can be readily calculated
by application of Henry's law.
412 VICTOR E. HALL
To determine the rate of oxygen consumption, the change of oxygen
content in any period was multiplied by the volume of water then
present, due allowance being made for the volumes removed in sampling.
These data are presented in Table III and are represented graphi-
cally in Fig. 3. It is clear that, under the conditions of these experi-
ments, the oxygen consumption of the worms decreased in an approxi-
mately linear manner with the oxygen pressure throughout the range
of 116.8 to 3.9 mm. Hg. It has already been noted that the oxygen
i "tisumption of worms active in U-tuhes bore no consistent relation to
oxygen pressure at least over the range of 138.2 to 93.3 mm. Hg.
These results are not necessarily in conflict, for (1 ) the ranges of pres-
sure are not the same, overlapping by some 25 mm. Hg, and (2) the
experimental conditions differed significantly.
Discussion
The large literature which has accumulated relative to the influence
of oxygen pressure on oxygen consumption has been most recently
reviewed by Helff and Stubblefield (1931), who list the animals studied
and classify their responses, and by Buchanan (1931), who gives a
short historical survey of theoretical interpretations of the relationship.
Various reasons have been given for the fall of oxygen consumption
which accompanies a decrease in oxygen pressure below a critical value.
The applicability of these suggestions to the respiratory mechanism of
Urcchis will be discussed :
i 1 ) ( ).\-yiien ilt-fieieaey in metabolizing cells due t« 'nurd equate tnins-
ptirt of o.vyrjcn to them. Accepting the doctrine of Pfluger (1872) that
the oxygen consumption rate of cells is determined by their own organ-
i/atinn and is independent of the concentration of oxygen in their
immediate milieu, provided that the latter be above zero, Kmgh (1916)
o mcluded that the decline in oxygen consumption with decreasing
oxygen pressure of the external medium was due to the attainment by
succe^ive Croups of cells of an oxygen-free state, with consequent
cessation of metabolism. This condition might result either from the
absence of adequate respiratory and circulatory mechanisms, or from
relatively slow diffusion of oxygen into the region of active oxidation.
This conclusion has been questioned by, among others, (lerard (1931),
who has shown by careful mathematical analysis of the interrelations
between the oxvgen consumption rate, pressure and diffusion rate in
the case "i" unicellular organisms that I 'finger's assumption is incom-
patible with experimental results, lie coin-hides that oxygen consump-
tion must change with oxygen pressure over a significant range in that
ACTIVITY AND O, CONSUMPTION OF URECHIS 413
region where the oxidation is actually taking place. Factors such as
alteration of permeability to oxygen or change in concentration of
oxidative enzymes (Buchanan, 1931), or decreased adsorption of oxygen
on catalysts of biological oxidation (Shoup, 1929) have been suggested
as possible mechanisms.
In Urechis, the respiratory and circulatory mechanisms, which have
been quantitatively analyzed by Redfield and Florkin (1931), are quite
effective, and possess large "factors of safety." The current of water
through the burrow carries about six times as much oxygen as is used
in metabolism. Only one-third of the oxygen taken into the hind-gut
is utilized. The attainment of equilibrium between hind-gut water and
blood, and between blood and active tissues, is facilitated by the peri-
staltic movement of both hind -gut and body- wall. The maximum dis-
tance from blood to muscle rarely exceeds one millimeter. The rate
of metabolism is such that only one-sixtieth of the blood oxygen content
is used per minute. The hemoglobin, fully saturated at the normal
physiological oxygen pressure, becomes an oxygen transporter at lower
oxygen pressures. From these considerations it seems justifiable to
conclude that deficient oxygen transport to the active cells is not respon-
sible for the fall in oxygen consumption with falling oxygen pressure
in the sea water provided to the animal.
(2) Accumulation of carbon dioxide might decrease oxygen con-
sumption, either of itself or by increasing the hydrogen ion concentra-
tion. In Experiment D, the three worms decreased the oxygen content
of the 2664 cc. of sea water by 11.0 cc. at the end of 4.35 hours, by
which time the oxygen pressure was 14.4 mm. and the oxygen consump-
tion reduced to 10 per cent of its initial value. Assuming an R. Q.
of 1, 11.0 cc. of carbon dioxide would be formed, which, in the volume
of 2664 cc., would increase the carbon dioxide concentration by
0.00017 M. From the data of McClendon (1917), it may be estimated
that this, in normal sea water, would cause an alteration of 0.05 mm.
Hg in carbon dioxide pressure, and a pH decrease of 0.07.
Moderate increases in CO2 pressure in the case of sea urchin eggs
(Warburg, 1910), the lobster Plomarus aincricaniis and the sand worm
Nereis wrens (Amberson, Mayerson and Scott, 1924), and certain
aquatic insects (Hiestand, 1931) did not decrease oxygen consumption.
Root (1930) found in fertilized Arbacia eggs that each 10 mm. Hg of
CO2 pressure reduced oxygen consumption by 21 per cent. It is ob-
vious that the change produced by 0.05 mm. CO2 pressure would be
negligible. Although Burfield (1928), using plaice eggs, and Fowler
( 1931), using Daphnia, found that CO., depresses oxygen consumption,
their experiments are not described in a manner permitting evaluation
of the small change under consideration here.
414 VICTOR E. HALL
It seems imi 'reliable that accumulation of carbon dioxide was a
major factor in the depression of oxygen consumption observed in
Urechis.2
(3) Alteration hi the intensity of muscular act'nnty. No careful
studies of the relation of the degree of spontaneous muscular activity
to the oxygen pressure of the external environment have been found
in the literature, although the necessity of controlling this factor is well
recognized. Attempts to remove its influence by anesthesia (Gaarder,
1918; F. G. Hall, 1929) introduce new complexities, as the anesthetic
used, ethyl urethane, is known to depress basic metabolism (Field and
Field, 1931).
As has already been stated, reduction to 70 mm. Hg of the oxygen
pressure of the water supplied to Urcchis in U-tubes produced con-
sistent changes in neither the degree of muscular activity nor the oxygen
consumption. On the contrary, the oxygen consumption did tend to
vary with the muscular activity. In this case, any influence which
l"\vered oxygen pressure may have had upon oxygen consumption was
overshadowed by the influence of muscular work. It is possible that,
had lower oxygen pressures been employed, an influence of this factor
might have been uncovered.
Unfortunately, no quantitative observations of the muscular activity
of the worms in jars were possible.
(4) Alteration in cliaraeter of metabolism. It is possible that the
exothermic processes yielding energy for basic and functional metabo-
lism, such as the decomposition of glycogen with the formation of lactic
acid, might proceed throughout the period at a relatively constant rate,
while the reconstitntive processes, which are directly or indirectly de-
pendent on oxidations involving molecular oxygen, might lag behind,
with the consequent accumulation of an "oxygen debt." In a single
experiment Table II 1. i Kxpcriment i ) some evidence of the occurrence
of such a process was obtained.
From the facts considered above it does not seem legitimate to
draw any positive com-lu>i<in-> as to the reason for the depression of
oxygen consumption accompanying the decreased oxygen content of
the water. The role of muscular activity and of the qualitative aspect
of metabolism merit further investigation.
2 Since Redfield and Florkin (1931) have shown that the oxygen dissociation
curve of Urcchis hemoglobin is not influenced by the carbon dioxide pressure,
criticism, such as Keys (1930) has urged against the work of F. G. Hall (1929)
and others on the grounds that carbon dioxide' would interfere with oxygen trans-
port, is inapplicable to the present investigation.
ACTIVITY AND O, CONSUMPTION OF URECHIS 415
SUMMARY
The greater part of the muscular activity of the echiuroid worm
Urechis caupo is involved in pumping a current of water through its
U-shaped burrow. The magnitude of this current was studied in
artificial burrows, food- free water being supplied to the animals. When
the animal is not feeding, the current amounts to about eleven cubic
centimeters per minute. During feeding periods, the rate of pumping
rises to about twenty-nine cubic centimeters per minute. The fre-
quency, duration and course of activity during these feeding periods has
been studied. The significance of the stream in relation to provision
of oxygen and food is discussed.
The oxygen consumption of the animals in U-tubes amounts to
0.00021 cc. per gram per minute, being comparable to that of related
forms. It is independent of the oxygen pressure down to a value of
70 mm. Hg.
The oxygen consumption of the worms when placed in covered jars
decreases with falling oxygen pressure throughout the range investigated,
115 to 4 mm. Hg. The reasons for this fall are discussed.
The author wishes to acknowledge his indebtedness to Mr. G. E.
MacGinitie for providing the animals used and for much useful advice
in their handling, to Dr. A. C. Redfield for suggestions which made
possible correlation of this work with that being carried out simultane-
ously by himself and Florkin on the same animal, and to Fr. L. Rudolph.
Mr. C. Watson and Mr. A. Fryer for assistance in carrying out the
experiments.
BIBLIOGRAPHY
AMBERSON, W. R., MAYERSON, H. S., AND SCOTT, W. J., 1924. Jour. Gen. Physiol.,
7: 171.
BUCHANAN, J. W., 1931. Biol. Bull, 60: 309.
BURFIELD, S. T., 1928. Brit. Jour. E.vper. Biol., 5: 177.
COHNHEIM, O., 1912. Zcitschr. ph\siol. Chcm., 76: 298.
FIELD, J., 2o, AND FIELD, S. M., 1931. Proc. Soc. Expcr. Biol. and, Med., 28: 995.
FISHER, W. K., AND MACGINITIE, G. E., 1928. Ann. Mag. Nat. Hist., Ser. 10, 1:
199 and 204.
FOWLER, J. R., 1931. Physiol. ZooL, 4: 214.
Fox, C. J. J., 1907. Pub. de Cir Constance. Copenhagen. No. 41.
GAARDER, T., 1918. Biochcm. Zeitschr., 89: 94.
GALTSOFF, P. S., 1928. Bull. Bur. Fish., 44: Document No. 1035.
GERARD, R. W., 1931. Biol. Bull, 60: 245.
iALL, F. G., 1929. Am. Jour. Physiol, 88: 212.
HELFF, O. M., AND STUBBLEFIELD, K. I.. 1931. Physiol. ZooL, 4: 271.
HIESTAND, W. A., 1931. Physiol. ZooL 4: 246.
KEYS, A. B., 1930. Bull. Scripts, hist., Tech. Ser., 2: 307.
KROGH, A., 1916. The Respiratory Exchange of Animals and Man. Longmans,
Green and Co. London.
LVSK, G., 1928. The Science of Nutrition, 4th Ed. Ch. IV. Saunders, Philadel-
phia.
416 VICTOR E. HALL
M. CLENDON, .1. P., 1917. Jour. Biol Chcm.. 30: 265.
MONTI OKI, A., 1913. Arch, ital Bio!., 59: 213.
,:. 1-:.. 1872. Pfliiger's Arch., 6: 43.
REDFIELD, A. C.. AM. FLORKIN, M., 1931. Biol. Bull., 61: 185.
ROGERS, C. G., 1927. Textbook of Comparative Physiology. McGraw-Hill Book
Co. New York.
K....T. W. S.. 1930. Biol. Bull., 59: 48.
SIK.I r. C. S.. 1929. Jour. Gen. Physiol.. 13: 27.
WARBURG, O., 1910. Zeitschr. physiol. Chcm., 66: 305.
WIXKLER, L. W.. 1888. Bcrichte Chcm. Gcs.. 21: 2843.
THE BLOOD PIGMENTS OF URECHIS CAUPO
J. P. BAUMBERGER AND L. MICHAELIS
(From the Jacques Loeb Laboratory, Hopkins Marine Station, Pacific Grove, Calif.)
The echiurid Urechis caupo was discovered by Fisher and Mac-
Ginitie and is an abundant inhabitant of the Monterey Bay in Cali-
fornia.1 One of its interesting features is its richness in hemoglobin.
This has been the subject of an extended study by Redfield and
Florkin.2 It is a peculiarity of this invertebrate that its hemoglobin is
contained within the blood cells and none in the blood fluid. Another
localization of the hemoglobin is the muscles, which are not vascular-
ized but contain all hemoglobin within the muscle cells; and, further-
more, the dorsal nerve chord appears red with hemoglobin. There are,
however, several other particular aspects with respect to the blood
pigment which are to be presented in this paper. In part, they are
concerned with changes, probably according to the age or to the seasons,
\vhich could not be fully studied during one season. The description
of these changes will be presented as they appeared to be and may be
subject to modifications as further studies may be extended over a
longer period of time.
The animals at our disposal varied in length, in the contracted state,
from 3 to 10 inches. Accordingly, the blood content of the body
cavity varied from 10 to 30 cubic centimeters. The color of the blood
varies, from the purest oxyhemoglobin-red to the darkest brown-black
or a black like Chinese ink, even after complete saturation with oxygen.
This variation of the blood is a very striking feature and obviously has a
definite physiological significance.
Red blood was encountered in some few of the smallest individuals,
and in some of the very largest sex-mature females. The majority of
the individuals, of medium size, contained brown or brown-black blood.
The blackest blood ever encountered was that of a very large sex-
mature male. The cause of the difference in color is revealed by a
microscopic examination. The red color of the blood is due to hemo-
globin homogeneously distributed within the blood cells. Whenever
the color is brown, besides this hemoglobin there is another, granular,
pigment of brown color within the cells which will be proved to be
1 Fisher, W. K., and MacGinitie, G. E., (1928), Ann. and Mag. Nat. Hist., Ser.
10, vol. 1, p. 199 and p. 204.
2 Redfield, E., and Florkin, M., 1931. Biol. Bull, 61: 185.
417
28
418 J. P. BAUMBERGER
hemutin. The description of the changes in these pigments may be
presented according to ideas developed during a study of two months.
This may not be sufficient to make sure of all details, and the whole
picture may be liable to some modifications upon more extended
studies.
\Ye start from a pure red blood in a young animal, recalling the
fact that not every small animal of our material contained the blood in
the red condition. In such an animal, the blood cells are spherical,
about 10-15 fji in diameter. The protoplasm is diffusely yellowish-
green with hemoglobin and, besides, rather tightly packed with color-
less granula of regular spherical shape, of a rather high refractory
index, — though not so high as that of fat drops,— and about 1 n in size.
No nucleus is visible in the fresh preparation but a nucleus becomes
\ isible after fixing and staining (fixed in acetone and stained with
-afranin). The nucleus is small, in the centre of the cell, and contains
.1 distinct nucleolus. Besides these cells, there is another kind, usually
somewhat smaller, much less numerous, containing yolk-yellow drop-
lets of a considerable size which often are conglomerated into a mul-
berry-like packet.
When the blood becomes brown, the granula of the hemoglobin-
containing cells are no longer colorless but are stained with a brown
pigment The granula, then, are no longer quite uniform in size and
spherical in shape, but somewhat more irregular. The size of the
cell is the same as in the red blood of young worms. This aspect was
most common among our material.
Now we come to the large sex-mature worms. Here a difference
arises according to the sex. One feature is common for both sexes.
The corpuscles become larger, up to 35 n in diameter, and m« >re \ ariable
in size. In the males, the brown pigment no longer stays exclusively
within the granula, but is more homogeneously scattered over the cell
so ih.it the hemoglobin color is overshadowed and can be detected only
by the spectroscope. The granula at the same time undergo a dis-
integration. They swell and have indistinct contours, being, as it
were, dispersed into a turbid mass without definite structure. At the
same time very small, spherical, quite black pigment granula, very
dense in structure, and not very numerous, are formed within the cell.
We do not know whether the development will go beyond this stage,
but it appears as though all hematin would gradually disappear and in
part be coiixerted into the dense black pigment.
The disintegration of the brown granula takes place in the females
also; but it does not lead to the formation of black granula within the
blood cells. Rather is the blood cell gradually deprived of any pigment
BLOOD PIGMENTS OF URECHIS
419
except for the hemoglobin. Instead, a pigment is formed within the
eggs, and there can be little doubt that the brown blood pigment is the
source of the black egg pigment.
The egg is a very large cell of almost the same aspect as that of
Asterias, also with respect to the size and shape of the nucleus. After
insemination, the nucleus disappears and the polar bodies are formed.
In the protoplasma of the egg a very fine dust of pigment granula is
scattered. The number of these granula is not very large so that the
eggs show macroscopically only a very slight yellowish-grey shade.
These pigment granula are in part black, in part somewhat more dark
red. The black pigment has the same shade as the one in the erythro-
cytes of the male, the difference being only that the black granula in the
eggs are usually smaller than those in the male erythrocytes. Upon
8
FIG. 1. Blood cell with regular, colorless granula of a relatively high refractory
power. Hemoglobin is diffusely dissolved in the protoplasma, not in the granula.
Fresh preparation.
FIG. 2. Blood cell with hematin-stained granule. Granula brown, protoplasma
yellowish-green with hemoglobin. Fresh preparation.
FIG. 3. The same, dry smear, fixed with acetone and stained with safranin.
Nucleus with nucleolus.
FIG. 4. Smaller cell with yolk-yellow droplets, containing sometimes also a few
hematin granula. Fresh preparation.
FIG. 5. Larger blood cell of a sex-mature female, with only a few hematin
granula. The whole protoplasma is diffusely yellowish-green with hemoglobin, with
no distinct structure. Fresh preparation.
FIG. 6. Larger blood of a sex-mature female, yellowish-green with hemoglobin,
without any hematin, very fine colorless granula of low refractory power. Fresh
preparation.
FIG. 7. The same, without any distinct granular structure, pure hemoglobin
shade in the whole protoplasma. Fresh preparation.
FIG. 8. The same, dry smear, fixed with acetone and stained with safranin,
showing the nucleus.
FIG. 9. Blood cell of an old male, black pigment besides colorless granula. The
protoplasma is diffusely light brown. Fresh preparation.
420 J. P. BAUMBERGER
confronting the fact that the black pigment is met, in the males, only
within the blood cells and never in the sperm, and in the females only in
the eggs and never in the blood cells, the interpretation seems unavoid-
able that the brown pigment is the mother material for the black one
and is utilized for the eggs in the female, but remains in the blood cells of
males.
It is likely that the brown pigment (which will be identified with
hematin) is converted, in part, into the black granular pigment, and
also in part into hemoglobin again. This latter conclusion is suggestive
because the sex-mature females with purely red blood have blood cells
of a much larger size than younger animals and yet these cells certainly
do not contain the hemoglobin in a lower concentration.
The blood cells can be hemolyzed by a copious amount of distilled
water, or in the undiluted blood, by some drops of ether, or better, by
gently shaking with a drop of octyl alcohol. The granula described
above will float isolated in a preparation of the laked blood. The color-
less granula remain as individuals, very often also the brown granula,
though these may also be disintegrated to finer pigment granula of
yellow brown color. All transitions can thus be observed from color-
less granula to partially and completely stained granula.
The chemical behavior of the hemoglobin has been fully described
by Red field and Florkin. It agrees in all its reactions and in all optic
properties with mammalian hemoglobin. It can be separated from
the brown pigment simply by centrifuging the blood hemolyzed with a
drop of octyl alcohol. The brown pigment is entirely insoluble and
forms the main part of the cake-like sediment, whereas the hemoglobin
is dissolved in the supernatant liquid. The brown pigment can be ex-
tracted from the cake-like sediment in the following way: The cake is
first extracted with acetone (or ether). A yolk-yellow pigment is here-
with extracted which is present either in the blood fluid or in the yolk-
yellow cells described above. \Yhen this extraction is complete, an-
other extraction is performed with acetone (or ether) containing acid
(glacial acetic acid or some drops of strong H(T). Hereupon the brown
pigment goes into solution and reveals the characteristic bands of acid
hematin. When this solution is reduced, either by shaking with solid
sodium hydrosulfite, or with platinum asbestos and hydrogen, and
pyridiiie is added, the characteristic spectrum of pyridine-hemochromo-
gen arises \\ith it> very distinct two bands even in highest dilution.
The broun pigment has herewith identified itself with hematin. The
pyridine-hemochromogen prepared from the hemoglobin, by treatment
with acid aceton, reduction and addition of pyridine, is spectroscopi-
cally identical with the one prepared in the same way from the hematin
BLOOD PIGMENTS OF URECHIS 421
granula. Both from the hemoglobin and from the brown granular
pigment Teichmann's crystals could be obtained.
It may be alluring to venture an interpretation of the physiological
significance of the changes occurring in the blood of this animal. We
prefer, however, to refrain from such an interpretation until experi-
ments of a more physiological nature are available.
ON THE RESPIRATORY FUNCTION OF THE BLOOD OF
THE SEA LION
MARCEL FLORKINi AND ALFRED C. REDFIELD
HOPKINS MARINE STATION, PACIFIC GROVE, CALIFORNIA
The capture of a Steller's sea lion, Eumetopias stcllcri, at the Hop-
kins Marine Station has afforded an opportunity to obtain certain data
on the conditions of equilibrium between the blood of an aquatic mam-
mal and the respiratory gases, which have not been available before.
The animal, which proved to be an old female, was blind, and having
been wounded with a rifle shot while it sat on the rocks in front of the
station, was secured with a gaff as it attempted to escape and brought to
shore. There it was killed by severing the great vessels in the neck
and a sample of 200 cc. of blood was collected as it flowed from the
wound. The animal was somewhat emaciated, but was not in a starving
condition as evidenced by a quantity of fish in its stomach and the
abundance of fat in the lacteals. The bullet wounds were found to be
limited to bony and muscular structures and had not caused extensive
bleeding. We are indebted to Dr. G. E. MacGinitie for placing the
blood at our disposal.
The blood was prevented from clotting by the addition of potassium
oxalate ; and was kept on ice during the subsequent sixteen hours in
which measurements were made. Samples were equilibrated with vari-
ous gas mixtures in a water bath at 38° C. for 20 minutes and then
analyzed for oxygen or carbon dioxide with the Van Slyke " constant
volume" apparatus. The gas mixtures were subsequently analyzed
with the Haldane gas analysis apparatus. The resulting data are re-
corded in Tables I and II. In order to correct the observed oxygen
contents for the dissolved oxygen, an absorption coefficient of a ==0.022
was assumed. The volume of erythrocytes in the blood was determined
with the centrifuge and proved to be 29 per cent of the total volume
of the blood.
Since all the observations recorded above were made in a short
period of time upon a single sample of blood, there was no opportunity
1 Fellow of the C. R. B. Educational Foundation.
422
RESPIRATORY FUNCTION OF SEA LION BLOOD
423
to check the results, which must in consequence be regarded as pro-
visional.
TABLE I
Data on the eqtiilibrium of sea lion's blood with oxygen. Temperature 38° C.
Carbon
dioxide
pressure
Oxygen
pressure
Oxygen
content
( ) \-ygen
dissolved
Oxygen as
oxyhemo-
globin
Saturation
mm. Hg
mm. Hg
vol.
per cent
vol.
per cent
vol.
per cent
per cent
42.20
27.53
5.23
0.08
5.15
25.9
42.80
32.00
6.75
0.09
6.66
33.5
47.00
43.00
10.95
0.12
10.83
54.5
26.20
14.32
2.12
0.04
2.08
10.5
21.20
34.55
10.70
0.10
10.60
53.4
24.20
27.55
12.75
0.08
12.67
64.8
24.30
33.30
11.56
0.10
11.46
57.7
106.50
61.55
11.67
0.18
11.49
58.0
air
air
20.40
0.45
19.95
100.5
air
air
20.21
0.45
19.76
99.5
TABLE II
Data on the equilibrium of sea lion's blood with carbon dioxide. Temperature 38° C.
Oxygen
pressure
Carbon dioxide
pressure
Carbon dioxide
content
Oxygenated
mm. Hg
150 ca.
150 ca.
150 ca.
vol. per cent
45.60
14.45
46.40
vol. per cent
39.15
22.45
38.01
Reduced
8.20
4.20
54.90
45.20
46.00
43.20
DISCUSSION OF RESULTS
In the blood of an aquatic mammal it is reasonable to look for con-
ditions which favor the circulation of oxygen to the muscles in order
to maintain the great energy expenditure required for rapid progression
through a viscous medium. One may also anticipate an increased
oxygen capacity to enable the animal to remain longer under water.
In the present instance the oxygen content of the blood when equi-
librated with air was 19.8 volumes per 100 volumes of blood. This
was not a greater oxygen capacity than commonly occurs in man and
other mammals. The volume occupied by the erythrocytes was only
29 per cent of the total blood, a figure much less than that commonly
424
A. C. REDFIELD AND M. FLORKIN
found in active terrestrial mammals. Each cubic centimeter of cor-
puscles combined with 0.68 cc. oxygen. Drastich (1928) has found
that in a large number of domestic mammals the concentration of hemo-
globin in the erythrocytes is approximately the same, being about 32
ims hemoglobin per 100 cc. blood corpuscles. Taking one gram of
hemoglobin tn combine with 1.34 cc. oxygen, each volume of corpuscles
i-ombines with 0.43 cc. oxygen. It appears then that the sea lion
corpuscles combine with about one and one-half times as much oxygen
as do those of the domestic mammals, i.e., the hemoglobin is just that
much more concentrated within them. We suspect that the blood under
examination may represent a somewhat anaemic condition and that the
blood of a younger and more vigorous sea lion would exhibit a higher
cell volume and oxygen capacity. Whether or not that is the case, there
can be little doubt that the unusual concentration of hemoglobin in the
0
< »xygen pressure
FIG. 1. Oxygen dissociation curves of blood of si a lion at 38° C. The ap-
proximate pressures of COa, in mm. Hg, at \vhieh the blood was equilibrated are
indicated by the numbers above the curves. Ordinate, percentage of saturation
with oxygen; abscissa, oxygen pressure in mm. Hg.
corpuscles of this specimen represents an advantageous condition in
that it minimizes the work which must be done by the heart in circulating
oxygen through the muscles.
Sud/.uki (1924) reports in the case of porpoise blood (Tummler-
RESPIRATORY FUNCTION OF SEA LION BLOOD
425
blut) oxygen capacities of 42.5 and 45.1 volumes per cent. The
erythrocyte count in the animals studied varied between 8.4 and 11.2
million per cubic millimeter. Since the erythrocytes of the Cetacea are
slightly larger than those of man (Marimoto, Takata, and Sudzuki,
1921), it would appear that in the porpoise the increased oxygen-carry-
ing power is accomplished by increasing the number of blood corpuscles
rather than by augmenting the concentration of hemoglobin within the
corpuscles.
1000
100
Si
O
u
10
0
10
50
60
70
20 30 40
Oxygen pressure at half saturation
FIG. 2. Oxygen pressures at which the blood of the dog, upper curve, and of
the sea lion, lower curve, are half saturated with oxygen in the presence of
varying quantities of CO2. Ordinate, pressures of CO2 in mm. Hg plotted on a
logarithmic scale; abscissa, oxygen pressure in mm. Hg at half saturation.
THE OXYGEN DISSOCIATION CURVE
In Fig. 1 curves are presented which indicate the general nature of
the equilibrium of sea lion blood with oxygen at various carbon dioxide
pressures. The general form and distribution of the curves resembles
that of the blood of other mammals. In order to compare equilibrium
conditions in the case of the sea lion with those characterizing the blood
of the dog, the pressures of oxygen at which the hemoglobin is half
saturated have been plotted in Fig. 2 against the corresponding carbon
426 A. C. REDFIELD AND M. FLORKIN
dioxide pressures. Similarly, a curve has been drawn representing this
relation in the case of the dog's blood from data kindly supplied by
Dr. D. B. Dill. It appears that oxygen is held at somewhat higher
tensions in the blood of the sea lion than in that of the dog. The dif-
ference between the two species is not greater than that exhibited by
various specimens of human blood, however. The advantage of this
di (Terence, in so far as it exists, in facilitating the rapid diffusion of
oxygen into the active muscle fibers, is obvious. The slope of the
curves also indicates that a given change in CCX tension will cause a
greater change in oxygen tension in the case of the sea lion, — again a
condition favoring the respiratory exchange.
THE CARBON DIOXIDE EQUILIBRIUM
The data in Table II serve to demonstrate the essential facts regard-
ing the equilibrium of carbon dioxide with the blood. If the data are
plotted, it will be found that the usual type of CO.2 dissociation curve
can be drawn through the points. The carbon dioxide combined at any
pressure is somewhat less than in the case of dogs studied in Dr. Dill's
laboratory. This condition may very probably be due to the presence
of lactic acid in the blood resulting from the struggles of the sea lion
in the course of its capture.
The difference in CO2 content of oxygenated and reduced blood is
similar to that of the blood of other mammals.
SUMMARY
The blood of -a sea lion, Eumetopias ^teller!, was found to have an
oxygen capacity of 19.8 volumes per cent.
The erythrocytes composed 29 per cent of its volume.
One volume of erythrocytes combined with 0.68 cc. oxygen, indi-
cating a hemoglobin concentration 50 per cent greater than that found
in domestic mammals.
The oxygen dissociation curves constructed at various pressures
conform to the usual mammalian type, but indicate that oxygen may be
held at slightly higher pressures than in the case of dog blood.
The carbon dioxide equilibrium is in no way remarkable and exhibits
the usual difference between oxygenated and reduced blood.
REFERENCES
I >:• ASTICII, L., 1928. Pfliirjcr's Arch.. 219: 227.
MARIMOTO, TAKATA AND SUDZUKI, M., 1921. Tohoku Jour. Expcr. Med., 2: 258.
SUD/I-KT, M., 1924. Tohoku Jour. ILrper. Mcd., 5: 419.
THE RESPIRATORY FUNCTION OF THE BLOOD
OF MARINE FISHES
R. W. ROOT
(From the Zoological Laboratory of Duke University, Durham, N. C.)
INTRODUCTION
The material embodied in this paper is a report of a study of marine
fish blood from the standpoint of respiratory function. Since we are
now fairly cognizant of the role of blood in mammals, it seemed to the
author that the scope of investigation should be widened by a study of
species other than mammals. In choosing marine fishes as experi-
mental material the writer had not only this point in mind, but, in
addition, the thought that fishes might present some new and interest-
ing aspect in blood physiology because of the fact that their method of
blood aeration is quite different from that of mammals. The blood of
mammals is apparently adjusted to the environment offered by the
alveoli of the lungs where high carbon dioxide tensions prevail and
oxygen tensions lower than in air exist. On the other hand, the gill of a
fish is bathed in a medium where higher oxygen tensions and much
lower carbon dioxide tensions prevail than is the case in the lung of a
mammal. In addition to these interesting differences, fish bloods
possess nucleated, instead of non-nucleated, red corpuscles, variable
quantities of hemoglobin (Hall and Gray, 1929), and function in vary-
ing, rather than constant, temperatures.
There is little work on the respiratory function of fish blood to be
found in the literature. Trendelenburg (1912), Gaarder (1918),
Krogh and Leitch (1919), Nicloux (1923), and Wastl (1928) have in-
vestigated the blood of fishes. Krogh and Leitch found a distinct
difference between oxygen dissociation curves for the bloods of the
fresh-water fishes, carp, pike, and eel, and the marine cod and plaice.
According to them, the hemoglobin of both types of fishes is very sensi-
tive to carbon dioxide, and the characteristics of their blood, as far as
the transportation of oxygen is concerned, are adjusted to the environ-
ment in which the fishes are living. Wastl has published oxygen dis-
sociation curves, carbon dioxide absorption curves, and figures for
arterial gas content and hydrogen ion concentration of carp blood.
Distinct differences were found between the blood of the carp and that
of mammals. Jolyet and Regnard (1877), and Kawamoto (1929) have
studied the blood of the eel. Kawamoto determined the relationship
427
l> R. W. ROOT
between the oxygen dissociation of the hemoglobin and temperature.
Collip (1920), Powers (1922), Jobes and Jewell (1927), and Kokubo
(1927, 1930) have investigated the alkaline reserve of several fishes.
Hall and collaborators (1926, 1928, 1929) have published data for the
hemoglobin concentration of the blood of a number of marine species.
The investigation to be reported in this paper has been restricted for
the most part to determinations of the oxygen capacities, oxygen disso-
ciation curves, carbon dioxide absorption curves, the effect of carbon
dioxide on the oxygen capacity, and the buffering capacities of the
bloods. The general results have been compared with similar results
obtained by other investigators on other vertebrates. The experi-
mental work was carried on at Woods Hole, Massachusetts, in the
laboratory of the United States Bureau of Fisheries.
METHODS
l-.xpcrinicntalAnimals. — The fishes that were employed in the study
are species common to the region of Woods Hole, Massachusetts.
Three species furnished most of the results, namely, the toadfish,
Opsanus tau (Linnaeus), the sea robin, Prionotus carolinus (Linnaeus),
and the common mackerel, Scomber scombrus (Linnaeus). Some work
was also done on the goosefish, Lophius piscatorius (Linnaeus), the
scup, Stenotomus chrysops (Linnaeus), and the puffer, Spheroides
maculatus (Bloch and Schneider). The fishes were maintained at the
laboratory under conditions as nearly normal as possible by keeping
them in "live-cars" or in hatching-boxes where plenty of running sea-
water was supplied at all times. The importance of keeping them in
good condition has been aptly pointed out by Hall, Gray, and Lep-
kovsky (1926).
The choice of the three fishes, the toadfish, sea robin, and mackerel
requires some explanation. Hall and Gray (1929), and Gray and Hall
(1930) have made a study of the blood sugar, hemoglobin, and iron of
these fishes and found a fairly precise correlation between these factors
and the activity of the fishes. The mackerel, for example, is an active
fish and is characterized by a high concentration of sugar, iron, and
hemoglobin in its blood, while the toadfish is a sluggish fish and is
characterized by a low concentration of blood sugar, hemoglobin, and
iron. The sea robin is more or less intermediate in this respect. On
the basis of this information it seemed worthwhile to broaden the
study enough to include several "type" fishes, instead of restricting ob-
servations to only one type. Another factor of a more practical turn
was influential in the choice of these fishes. The blood of fishes does
RESPIRATORY FUNCTION OF BLOOD OF MARINE FISHES 429
not lend itself easily to gas analysis. This has been recognized by
others, and is probably one reason why more work has not been done.
On account of the small size of many fishes, blood is not easily obtained
for study. Some fishes have very fragile red corpuscles which makes it
almost impossible to subject their blood to the drastic treatment
necessary in determining dissociation curves. Also fish blood reacts
peculiarly toward the reagent, potassium ferricyanide, used to liberate
oxygen. As soon as the reagent comes in contact with the blood a
coagulum is formed. Under these conditions it is quite impossible to
liberate all the oxygen from the blood without subjecting it to vigorous,
prolonged shaking. The blood from the fishes employed reacts no
differently from other fish bloods toward ferricyanide, but is quite suit-
able in other respects. This is especially true of toadfish and sea robin
bloods. Mackerel blood is quite viscous and makes pipetting rather
annoying. It is also the hardest of the three to handle in the Van
Slyke extraction chamber, for its coagulum adheres to the walls and is
not easily cleaned out.
Obtaining of Blood Samples. — In obtaining blood for analysis an
attempt was made to standardize conditions as much as possible.
When it was not desired to know the gas content actually existing in the
blood at the time of drawing, the procedure was to remove a fish quickly
from the water and bleed it from the gills by means of a hypodermic
needle attached to a 5 or 10 cc. syringe. Lithium oxalate was used as
an anticoagulant. The time of bleeding was made as short as possible
in order to avoid getting blood that might have excess acid in it on ac-
count of asphyxial conditions. Hall (1928) has shown that asphyxia in
fishes lowers the oxygen capacity of their blood considerably.
Since most of the fishes used were small, it was found necessary to
combine the blood of several specimens of a species. This practice led
to no ill effects. In fact, the analytical results on different blood speci-
mens checked more closely than otherwise on account of the averaging
effect of such a procedure.
The blood was used as soon as it was drawn. In preliminary work
addition of both sodium fluoride and potassium cyanide to the blood to
prevent respiration of the cells and loss in carbon dioxide-combining
power was tried. The results were unsatisfactory. The slight loss in
carbon dioxide-combining power over a period of time did not appear
to be checked. Rather than add more extraneous chemical factors, it
was finally decided to modify the procedure in such a way as to avoid
any appreciable error due to the activity of the cells. This necessitated
using a given sample of blood a shorter length of time and checking a
curve that had once been established by means of freshly drawn blood.
430 R. W. ROOT
It also made it necessary that a blood sample be analyzed for its gas
content as soon as it had come into equilibrium with a given gas tension,
and that the gas phase be separated from the blood remaining in the
tonometer during the time consumed in the analysis. It should be
mentioned at this time that Dr. F. G. Hall (unpublished) has deter-
mined the oxygen consumption of these bloods and shown, under the
conditions of the author's technique, that the error arising from oxygen
consumption of the cells would be negligible over the short period of
time that elapses in getting a blood sample into the Van Slyke appara-
tus from the tonometer.
\Yhen it was desired to know the actual content of gases existing in
the blood at the time of drawing, the method was modified to suit the
purpose. In attempts to determine arterial or venous gas contents,
fishes were placed in suitable traps and a stream of fresh sea-water
directed over their gills. The blood was then drawn under oil and the
gases immediately analyzed. It is most difficult to get a satisfactory
technique for determining arterial and venous gases in fishes. The
results obtained are only approximate at best.
Determination of Erythnx ytc Count and Volume. — The number of red
corpuscles per cubic millimeter of blood was determined by employing
the usual procedure. The volume of red corpuscles was determined by
an haematocrit especially designed by Dr. F. G. Hall for use with fish
blood.
Equilibration of Blood u'ith Gases and Determination of Gases. — The
gases used in these experiments were carbon dioxide, oxygen, and ni-
trogen. The required mixtures were made in a mixing chamber at-
tached to an ordinary gas burette (if gas mixtures different from air
were desired). The method of handling the blood and gases was essen-
tially the same as that prescribed by Austin et al. (1922), except for the
admittance of gases to tonometers. Instead of using the method they
prescribe, the tonometers were filled with clean, neutral mercury, and
the gas mixtures drawn into them from the mixing chamber by with-
drawing the mercury. The equilibration of blood samples was carried
out according to their "first saturation method," using the double
tonometer. Equilibration for all samples was allowed to take place at
20° C. and at atmospheric pressure. Atmospheric pressure was main-
tained by occasionally opening the stop-cock on the tonometer. Since
the gases in the tonometer were always analyzed after equilibration, the
entrance of a small amount of gas from the atmosphere did no harm.
The tonometers were mechanically rotated in a thermostatically con-
trolled water bath for a period of about 30 minutes. It was found in
preliminary experiments that this was sufficient time to allow the blood
RESPIRATORY FUNCTION OF BLOOD OF MARINE FISHES 431
and gas phase to come into equilibrium with each other. Usually one
tonometer was rotated at a time. However, in some of the work in-
volving carbon dioxide absorption, two tonometers were used simul-
taneously, one containing reduced and the other oxygenated blood.
At the end of equilibration a sample of blood was removed from the
tonometer and the gases in it immediately analyzed according to the
technique of Van Slyke and Neill (1924). Both oxygen and carbon
dioxide were simultaneously liberated from the blood by using acid
ferricyanide. One cubic milliliter of blood was used for each analysis,
and was admitted to the extraction chamber by means of a Van Slyke
differential pipette. The blood was agitated in the extraction chamber
a little longer than is usual for mammalian blood. This was found
necessary in order to insure the complete liberation of gases. Both the
carbon dioxide and oxygen were absorbed after liberation, sodium
hydroxide being used for carbon dioxide, and sodium hydrosulfide for
oxygen.
TABLE I
Oxygen capacity determinations. Blood equilibrated with air at 20° C.
Red
Species
Oxygen capacity
Blood
Haematocrit
Iron
Corpuscles
•vol. per cent
cu. mm.
vol. per cent
mg. 100 cc.
Goosefish
5.07
867,083
15.45
13.40
Toadfish
6.21
585,000
19.50
14.00
Puffer
6.75
2,284,000
17.50
21.50
Scup
7.30
2,685,000
32.60
24.60
Sea robin
7.66
2,536,000
24.00
23.10
Mackerel
15.77
3,000,000
37.10
37.10
The amount of carbon dioxide and oxygen in blood was expressed as
volumes per cent of dry gas at 760 mm., and 0° C., the tables prepared
by Van Slyke and Neill (1924) being used for oxygen, and those pre-
pared by Van Slyke and Sendroy (1927) for carbon dioxide. In deter-
mining the oxygen combined with hemoglobin, the amount of oxygen
physically dissolved was calculated on the basis of Bohr's (1905)
solubility coefficients. A special equation similar to that of Peters,
Bulger, and Eisenman (1923) was employed in the calculation to allow
for the variable corpuscular volume in the various bloods.
The concentration of the gaseous phase in the tonometers was deter-
mined after equilibration of blood samples by analysis in the Haldane
apparatus as modified by Henderson (1918). The results were ex-
pressed in terms of tension by employing the usual calculations.
Method of Studying Lactic Acid Effect. — Lactic acid was carefully
added to small samples of blood in amounts necessary to give the de-
432
R. W. ROOT
sired concentration. The blood was then equilibrated in air and
hamlk-d the same as in the other experiments.
Calculation of pll of Blood. — In calculating the pH of fish blood the
familiar Henderson-Hasselbalch equation was used (Henderson, 1908;
Hasselbalch, 1917). A pK' factor of 6.24 was employed for the blood
at 20° C. This was derived by using the average pK' factor of 6.13
that has been worked out for mammalian serum at 38° C. (using Bohr's,
1905, solubility coefficient for CO«) by a series of workers (Warburg,
1922; Cullen, Keeler, and Robinson, 1925; Van Slyke, Hastings, Mur-
ray, and Sendroy, 1925; and Hastings, Sendroy and Van Slyke, 1928),
and adding a temperature correction of 0.005 for each degree below
100
10 20 jo 40 so 60 ro ao 90 100
Pa m m. Kg
FIG. 1. Oxygen dissociation curves for toadfish blood at 20° C. Curve 1 at
1 mm. carbon dioxide; curve 2 at 10 mm. carbon dioxide; and curve 3 at 25 mm.
carbon dioxide tension.
38° C. (Hasselbalch, 1917; and Warburg, 1922). In addition a correc-
tion of 0.02 was added 1 it-cause whole blood was used instead of serum.
The pK' factor for whole blood is slightly higher than that for serum
(Warburg, 1922; Peters, Bulger, and Eisenman, 1923; and Van Slyke
etal., 1925).
In using the pK' factor in the following calculations of pH, it is
recognized that there are many variables which enter into its composi-
RESPIRATORY FUNCTION OF BLOOD OF MARINE FISHES 433
tion for any one blood, especially when it is applied to whole blood.
Warburg (1922), Hastings and Sendroy (1925), Stadie and Hawes
(1928), and Stadie (1928) have shown that the pK' factor is affected by
the ionic strength of the solution in which it is measured. Further-
more, the researches of Warburg (1922), Van Slyke, Wu and McLean
(1923), and Peters, Bulger, and Eisenman (1923) have demonstrated
the effect of degree of oxygenation of blood, its pH, and its relative
volume of corpuscles and plasma upon the pK' factor.
100
9o
70
50
1O
_Q
O
10
•40
t>o
TO
60
10O
Curve 1 at
FIG. 2. Oxygen dissociation curves for sea robin blood at 20° C.
1 mm. carbon dioxide; and curve 2 at 25 mm. carbon dioxide tension.
However, there is little information at the present time that will
permit the calculation of the pH of fish blood with the degree of refine-
ment that now seems possible for mammalian blood. Therefore the
author does not claim absolute accuracy for the calculated pH of fish
blood, but only relative, and admits that with the advent of more in-
formation his figures will probably require correction.
RESULTS
A . Tine Transportation of Oxygen
Oxygen Capacity of Blood. — The results of this study are summar-
ized in Table I. The figures for oxygen capacity are those obtained
when the blood was equilibrated in air, and dissolved oxygen sub-
tracted. Thus they represent the actual amount of oxygen combined
29
4.U
R. W. ROOT
with hemoglobin under the conditions of the experiment. An attempt
has been made to correlate oxygen capacities of the various bloods with
their corpuscle count, corpuscle volume, and iron content. The author
i- indebted to Dr. F. G. Hall and Mr. S. R. Tipton for some of the data
contained in the last three columns of Table I. It should be mentioned
TABLE II
Gases in Blood, as Drau*n tinder Oil
Species
Kiiul
of
Blood
COj
1 >S
Pco:
Pos
111,1 )
Conditions of Drawing
vol.
vol.
m m .
//,'.
III III .
11,
per
, , nt
Scup
Arterial
8.15
8.33
'
Water over gills
Blood from gills
9.16
5.00
—
—
59
Water over gills
Blood from caudal artery
8.90
5.24
— •
—
69
Water over gills
Blood from gills
11.40
3.13
—
—
—
Water over gills
Blood from gills
Sea robin
6.15
2.55
2
10
33.2
Water over gills
Blood from gills
Toadfish
\Vnous
13.30
0.54
10
2
7.6
Water over gills
Blood from heart
• M-flsh
10.25
[*rac(
—
—
—
W.Mer D\ rr trills
Blood from bulbus
Puffer
1 L90
0.34
—
—
5.3
Water over gills
lUond from sinus venosus
Sea robin
Aspliyxial
2.41
4
20
31.5
1 i-lt in air
1 U(H id from gills
13.40
L.59
Id
20
22.2
1 i-h in air
I'lixj.l tri nn gills
t hut the figures for corpuscle count, corpuscle volume, and iron content
u ere not always obtained from the same samples of blood on which
oxygen rup.irity determinations were made. The data represent the
average ol .1 < on-iderable number of determinations. There appears
to be a ;jrnrial correlation between the oxygen capacity of fish blood
and the previously mentioned factors. The best agreement exists be-
tween iron and oxygen. Since the corpuscle count and volume are
\ariable anion^ themselves, on account of differences in size of cor-
puscles, these factor-, do not show as good a correlation as iron.
RESPIRATORY FUNCTION OF BLOOD OF MARINE FISHES 435
The most interesting feature of this phase of the work is that it
points out great differences in the oxygen capacities of the various
bloods. The sluggish goosefish and toadfish possess bloods of low
oxygen capacity, whereas the active mackerel has a blood of high
oxygen capacity.
TABLE III
Oxygen dissociation of blood. Equilibrated at 20° C.
Species
PC02
P02
O2-Ca-
pacity
O2-Con-
tent
O2-Dis-
solved
Oz Com-
bined
HbO2
PH
mm. Hg
mm. Hg
vol.
vol.
vol.
vol.
per cent
per cent
per cent
per cent
per cent
Toadfish
0.762
3.75
6.84
1.56
0.015
1.54
22.5
7.86
0.454
7.80
5.13
1.75
0.030
1.72
33.5
7.99
1.150
10.30
6.31
2.66
0.040
2.62
41.6
7.68
0.615
24.60
5.13
3.78
0.096
3.68
71.8
7.60
0.765
39.20
5.13
4.70
0.152
4.55
88.3
7.78
0.690
56.20
5.13
5.28
0.219
5.06
98.6
7.70
0.690
80.00
5.13
5.51
0.312
5.20
101.4
7.66
8.62
8.85
6.31
1.65
0.035
1.62
25.6
7.33
10.85
13.35
6.68
2.37
0.052
2.32
34.6
7.21
11.15
35.70
6.68
3.64
0.138
3.50
52.3
7.18
11.25
48.20
6.68
4.00
0.188
3.81
57.0
7.16
10.42
80.00
6.68
4.75
0.312
4.44
66.4
7.17
10.28
103.00
6.68
5.25
0.400
4.85
72.6
7.16
25.40
7.47
5.56
1.63
0.029
1.60
28.8
6.98
25.10
9.63
6.31
1.43
0.037
1.37
22.0
7.00
27.40
10.00
6.84
1.93
0.039
1.89
27.5
6.94
25.80
53.00
5.56
2.62
0.207
2.41
43.5
6.97
25.05
100.00
5.56
3.38
0.390
2.99
53.8
6.98
Sea robin
0.304
6.69
7.02
2.15
0.026
2.12
30.2
8.22
0.485
19.40
7.80
4.09
0.075
4.02
52.1
8.16
0.227
27.00
8.20
6.72
0.100
6.62
80.7
8.03
0.727
39.70
6.82
5.47
0.155
5.32
78.0
7.86
1.050
47.20
7.91
6.70
0.184
6.52
82.5
7.67
0.455
54.50
8.20
7.55
0.212
7.34
89.5
7.67
1.510
59.70
7.25
6.40
0.233
6.17
85.1
7.68
1.160
79.00
7.25
7.22
0.308
6.91
95.3
7.43
1.132
99.00
7.91
8.19
0.386
7.80
98.6
7.13
0.761
104.50
7.04
6.94
0.408
6.53
93.0
7.83
0.225
109.00
7.20
7.50
0.425
7.08
98.3
8.09
24.70
9.35
7.66
0.42
0.036
0.38
6.3
7.05
28.10
10.75
7.10
0.32
0.042
0.28
4.0
6.98
25.00
13.50
6.85
0.50
0.053
0.45
6.5
7.03
26.10
17.00
7.15
0.85
0.066
0.78
11.0
7.00
24.40
17.80
6.97
0.89
0.069
0.82
11.6
7.04
21.80
18.20
7.15
0.83
0.071
0.76
10.8
7.13
25.10
42.00
7.15
1.61
0.164
1.45
20.2
7.06
25.60
60.50
7.15
1.94
0.236
1.70
23.8
7.04
26.00
84.50
6.97
2.12
0.330
1.79
25.6
7.05
25.90
92.50
7.15
2.50
0.360
2.14
30.0
7.00
23.60
106.50
6.97
2.36
0.415
1.95
28.0
7.12
436
R. W. ROOT
TAIU !•: 1 1 {—Continued
Pcoj
Po2
02-Ca-
pacity
Os-Con-
tent
O«-Dis-
solved
Oi Com-
bined
HbOi
pH
mm. Hg
mm. Hg
ml.
per cent
vol.
per cent
per cent
vol.
per ,fiit
per cent
.M.i<-k( ivl
1.130
4.74
16.41
1.76
0.018
1.74
11.0
8.19
1.250
8.30
15.76
3.46
0.032
3.43
21.6
8.00
0.640
17.7(1
14.72
8.11
0.067
8.04
53.9
S.I 7
. 0.977
31.95
16.29
12.45
0.122
12.33
75.8
7.96
0.754
45.00
16.64
13.60
0.172
13.43
80.7
8.03
0.382
64.60
17.81
16.35
0.247
16.10
90.43
8.00
0.825
75.60
16.64
15.30
0.290
15.01
90.2
7.86
0.768
98.70
15.59
14.75
0.378
14.37
92.2
7.48
0.758
115.00
10.64
16.10
0.440
15.66
94.1
7.60
10.17
9.87
14.54
1.01
0.038
0.97
6.7
7.49
10.10
14.50
14.54
2.15
0.055
2.10
14.5
7.58
1 1 .30
29.40
14.54
4.13
0.113
4.02
27.7
7.51
1(1.15
46.80
14.54
7.15
0.179
6.97
48.0
7.51
1 1 .00
65.00
14.54
7.7(1
0.249
7.45
52.0
7.49
10.30
77.50
14.54
<>.S2
0.297
9.52
65.5
7.43
10.35
83.30
14.54
10.30
0.319
9.98
68.6
7.46
10.85
101.00
14.54
11.00
0.387
10.61
73.0
7.42
24.50
12.00
17.81
0.92
0.046
0.87
5.0
7.33
18.85
40.50
16.62
6.35
0.155
6.20
37.3
7.36
24.80
53.40
16.62
7.25
0.204
7.05
42.5
7.25
24.80
o<).50
16.62
8.65
0.267
8.38
50.5
7.16
25.40
70.30
16.62
9.50
0.269
9.23
55.6
7.14
21.SH
90.00
10.02
11.08
0.345
10.74
64.6
7.14
19.60
101.50
17. SI
9.58
•i ;s<)
9.19
51.7
7.33
23.90
115.20
16.02
12.70
0.442
12.26
73.8
7.11
Oxygen Content of ttlood. — Any attempt to determine the actual
amount of gas existing in the arterial or venous blood of fishes as small
as those used in this investigation is beset with difficulties. The few
results obtained are recorded in Table II. Attempts to get arterial
blood from these fishes were rewarded with little success. Analysis of
blood removed from efferent gill arteries of the scup and sea robin
-bowed much less oxygen than could reasonably be expected. It would
appear that the syringe used in the operation hastened the circulation
through the gill to a point where the blood had not sufficient time to be-
< ome a< rated to the normal degree. A more reliable source of arterial
blood i> that from the caudal artery, but th<- M-hes used are unsuited for
getting blood from such a source. Until a more adequate technique is
devised, any statement as to the actual oxygen content of arterial blood
in these fishes will have to be postponed. 1 lall ( 1()30) reported 85 per
cent oxygen saturation in mackerel arterial blood. Wastl (1928) found
''^ | XT (cut oxygen saturation in carp blood.
RESPIRATORY FUNCTION OF BLOOD OF MARINE FISHES 437
With regard to the oxygen content of venous blood, more satis-
factory results were obtained. Practically no oxygen was found in the
venous blood of the goosefish, toadfish, and puffer.
The gas tensions recorded in Table II for sea robin and toadfish
bloods were not determined experimentally but were interpolated from
the oxygen dissociation curves for their bloods.
Oxygen Dissociation of Hemoglobin. — Table III, and Figs. 1,2, and 3
summarize the results of this study. At a carbon dioxide tension of
so 60 70 so 90 too no 120
10 20
FIG. 3. Oxygen dissociation curves for mackerel blood at 20° C. Curve 1 at
1 mm. carbon dioxide; curve 2 at 10 mm. carbon dioxide; and curve 3 at 25 mm. car-
bon dioxide tension.
approximately one millimeter toadfish hemoglobin is characterized by a
steeper dissociation curve than either sea robin or mackerel. The
hemoglobins of the latter appear to act quite the same toward oxygen,
at this carbon dioxide tension, except for the fact that sea robin hemo-
globin tends to become saturated a little more quickly than mackerel
at the higher oxygen tensions. At 10 mm. carbon dioxide tension the
dissociation curves for mackerel and toadfish hemoglobins are flattened
most remarkably. A still more pronounced flattening is produced at
25 mm. carbon dioxide tension. Of the three hemoglobins the sea
438
R. W. ROOT
robin's is most affected at the latter carbon dioxide tension. The ap-
pearance of the curves at 10 and 25 mm. of carbon dioxide is very
interesting. There is a tendency for them to become nearly asymptotic
\vith respect to the abscissa before saturation is complete. This is most
noticeable in the curves for toadfish and sea robin hemoglobins. At
10 mm. carbon dioxide tension, in the case of toadfish hemoglobin, the
mrnJiy
90 IOC J10
FK.. 4. KflVct of carbon dioxide on the oxygen capacity at 20° C. Curve 1 is
for sea robin blood; curve 2 for mackerel blood; and curve 3 for toadfish blood.
curve appears to approach a limit at approximately 75 per cent oxygen
saturation. At 25 mm. carbon dioxide this same tendency occurs in
sea robin hemoglobin at 25 per cent saturation, while toadfish hemoglo-
bin shows this at about 50 per cent saturation. The curves for mack-
erel hemoglobin do not show any very marked tendency to become
asymptotic. In the case of toadfish and sea robin bloods it would ap-
pear as if carbon dioxide affected not only the 0x3 ^en dissociation con-
stant of the hemoglobin, but, also, that the quantity of oxygen with
which the hemoglobin can combine is reduced by the presence of carbon
dioxide. Red field and Mason (1928) have pointed out that such an
effect is produced by acid in the case of purified Limulns hemocyanin.
l-.tl'i-il of Carbon Dioxide on the Oxygen Cajxn'ily. — The peculiar
Hl'ects of carbon dioxide on the oxygen dissociation curves suggested an
invt --li'j.iiion of its effect on the so-called oxygen capacity. For this
\\ork blood samples were equilibrated with 15S nun. of oxygen and
RESPIRATORY FUNCTION OF BLOOD OF MARINE FISHES 439
varying tensions of carbon dioxide. The results are presented in
Table IV, and Figs. 4 and 5. As can be seen from the data, carbon
TABLE IV
Effect of carbon dioxide on oxygen capacity. Blood equilibrated at constant
Po2(152 mm.) at 20° C.
Species
PC02
O2- Ca-
pacity
O2-Con-
tcnt
O2 Dis-
solved
O2 Com-
bined
Loss in
oxygen
apacity
pn
mm. Hg
vol. per cent
vol.
vol.
vol.
per cent
at O Pro,
per cent
per cent
per cent
Toad fish
1.37
6.87
7.22
0.60
6.62
3.64
7.64
1.54
6.34
6.67
6.07
4.27
7.52
4.62
6.34
6.50
5.90
6.94
7.42
9.15
6.34
5.50
4.90
22.70
7.23
11.75
6.40
5.03
4.43
30.80
7.20
16.70
6.34
4.38
3.78
40.40
7.08
25.70
6.34
3.85
3.25
48.80
6.98
29.30
5.59
3.60
3.00
46.40
6.94
41.00
6.63
3.98
3.38
49.20
6.82
63.70
6.40
3.57
2.97
53.60
6.71
88.00
6.40
3.50
2.90
54.70
6.60
106.00
6.87
3.58
2.98
56.60
6.48
Sea robin
1.21
7.67
7.92
0.60
7.32
4.50
7.79
3.79
7.67
7.15
6.55
14.60
7.56
10.50
7.71
4.95
4.35
43.60
7.30
12.15
7.67
5.06
4.46
41.80
7.27
15.25
7.71
4.53
3.93
49.00
7.11
25.05
7.69
3.67
3.07
60.00
7.06
43.60
7.15
3.20
2.60
63.60
6.83
53.00
7.67
3.26
2.66
65.40
6.75
80.80
8.29
3.04
2.44
70.60
6.68
102.00
7.15
3.06
2.46
65.60
6.60
103.00
8.70
3.34
2.74
68.60
6.58
106.50
7.15
2.82
2.22
69.00
6.57
107.00
8.15
2.90
2.30
71.80
6.56
Mackerel
2.17
16.43
16.80
0.586
16.21
1.34
7.94
2.26
16.78
17.15
16.56
1.31
7.84
12.00
15.64
14.75
14.16
9.90
7.37
22.50
16.64
14.05
13.46
19.10
7.08
26.10
14.51
11.05
10.46
27.90
7.21
37.30
16.78
9.40
8.81
47.50
7.10
40.80
16.43
9.81
9.22
43.80
7.16
65.05
16.60
8.35
7.76
53.25
6.95
80.00
16.78
6.90
6.31
62.60
6.88
95.00
16.78
7.06
6.47
61.60
6.84
108.50
16.43
6.96
6.37
61.20
6.82
dioxide affects a very marked loss in oxygen-combining power of the
hemoglobins. However, a maximum loss is reached beyond which
further addition of carbon dioxide has little or no effect. Sea robin
440
R. W. ROOT
hemoglobin suffers the greatest loss in oxygen-combining power, the
maximum being around 70 per cent, whereas the maximum for toadfish
i- about 55 per cent. Mackerel hemoglobin has a maximum loss be-
tween those for the other two. The data procured seem to corroborate
what uas already anticipated in a study of the dissociation curves,
namely that the ability of the hemoglobins to combine with oxygen is
great ly reduced in the presence of carbon dioxide.
ao
65O 6/9" 7OO 725 7 SO 77S 6 CO
P*
\:K',. 5. Effect of pH on the oxygen capacity at 20° C. Curve 1 is for sea robin
blood; curve 2 for mackerel blood; and curve 3 for toadfish blood.
Plotting loss in oxygen capacity, or, as designated in Fig. 5, loss in
oxyhemoglobin, as a function of pH, yields sigmoid curves for the three
hemoglobins. \Yithin a certain range of pi I there is a marked loss in
oxygen-combining pourr. ( )utside this range at either end, within the
limits of pH established in these experiments, loss in oxygen-combining
power is relatively slight.
I'.ffect oj 'Lactic Acid on Oxygen Capacity. — It was t hough t advisable
to modify the pH of the bloods by other means than the use of carbon
dioxide and 'see if a similar effect on the oxygen capacity could be ob-
t, lined. Therefore blood samples containing definite concentrations of
lactic acid were equilibrated in 15S mm. ot oxygen. In this case, of
i ourse, no carbon dioxide was added to the gaseous phase in the tonom-
eters. ( )nly the blood of the sea robin was used in these experiments.
The results are shown in Fig. 6. A greater loss of oxyhemoglobin was
observed at the higher concentrations of lactic acid than was found
RESPIRATORY FUNCTION OF BLOOD OF MARINE FISHES 441
when carbon dioxide was used, though the calculated pH was less.
However, there may have been some other factor entering in to produce
the results, such as the formation of methemoglobin, and, since this
was not ascertained, no emphasis should be placed on the magnitude
of the results. The main matter of interest is that, in general, the form
of the curve is similar to that for the carbon dioxide effect.
too
do
I
a>
/o
O
.Of
• OZ
.03
.04
M Lactic acid
FIG. 6. Effect of lactic acid on the oxygen capacity at 20° C.
robin blood only.
Curve for sea
B. The Transportation of Carbon Dioxide
Carbon Dioxide Content of Blood. — An attempt was made to deter-
mine the amount of carbon dioxide normally present in the circulating
blood. The results are recorded in Table II. The bloods of the fishes
studied contain relatively little carbon dioxide. The tension even in
the venous blood is probably not more than 10 to 15 millimeters.
Carbon Dioxide Absorption of Blood. — The results of this study are
presented in Figs. 7, 8, and 9. Of the three bloods examined the toad-
fish was found to take up the least, and the mackerel to take up the
most carbon dioxide. All three curves tend to flatten out above 10
mm. carbon dioxide tension, the flattening being most pronounced in
the case of toadfish blood, and least in mackerel. The curve for
mackerel blood is quite out of the class of the curves for the other two
442
R. W. ROOT
fislu - Apparently mackerel blood has a higher available base than
toadtish and sea robin bloods.
Christiansen, Douglas, and Haldane (1914) were the first to dis-
cover that reduced blood will take up more carbon dioxide than oxy-
genated blood. This phenomenon has been explained since their
\\ork was published by the assumption that oxyhemoglobin is a
stronger acid than hemoglobin, and, thus, base is liberated and made
a\ailal>le for carbon dioxide when oxyhemoglobin is reduced. The
elucidation of the fact is due mainly to the work of Van Slyke and his
collaborators at the Rockefeller Institute.
JO 2O JO 4O SO bo 7O QO 9O
tfO
FIG. 7. Carbon dioxide-absorption curves for toadfish blood at 20° C. The
dots are for reduced, and the circles for oxygenated blood.
Since it has been demonstrated beyond doubt that reduced blood
will take up more carbon dioxide than oxygenated, as far as mammals
are concerned, it was thought advisable to determine whether a similar
phenomenon could be shown for fish blood. \Yastl (1928) has shown
such to be the case as far as carp blood is concerned. The results ob-
laincd on the Moods of the toadfish, sea robin, and mackerel are shown
in the carbon dioxide-absorption curves drawn in Figs. 7, 8, and 9.
'I »,i<llish and sea robin bloods show little ditlerrnce in the ability of
reduced and oxygenated to absorb carbon dioxide. \Yithin what ap-
pears to be the physiological range of carbon dioxide tension (from
analyses of the carbon dioxide content of \enous blood), however, re-
duced blood takes up slightly more carbon dioxide than oxygenated.
\\ ith respect to mackerel blood the range where this can be demon-
strated is considerably greater, and the curves begin to take on the ap-
pearance of mammalian carbon dioxide-absorption curves. There are
RESPIRATORY FUNCTION OF BLOOD OF MARINE FISHES 443
probably at least two reasons why it is difficult to demonstrate greater
carbon dioxide absorption by reduced than by oxygenated blood in the
case of the first two fishes: (1) the small amount of hemoglobin present
to furnish base in changing from the oxygenated to the reduced state,
and (2), the effect of carbon dioxide in reducing the oxygen capacity.
One can hardly say he is dealing with oxygenated blood at high carbon
dioxide tensions, for under these conditions the oxygenation of the
blood is greatly reduced.
JO
FIG. 8. Carbon dioxide-absorption curves for sea robin blood at 20° C. The
dots are for reduced, and the circles for oxygenated blood.
Buffering Ability of Blood. — The BHCO3 concentrations of the
bloods have been calculated and the results plotted as a function of pH.
Such a procedure will point out their relative buffering ability. The
curves obtained are shown in Figs. 10, 11, and 12. In general, within
the normal range of pH, reduced blood has a higher concentration of
BHCOs at a given pH than oxygenated. This means that by reduction
oxyhemoglobin imparts to the blood a certain protection against change
in pH, for a certain added amount of carbon dioxide may be taken up
at the same hydrogen ion concentration. Outside the normal pH
range the curves for oxygenated and reduced blood tend to converge so
that there is practically no difference in the ability of the two states of
blood to bind carbon dioxide.
A comparison of the three bloods shows at once that mackerel blood
is much better buffered than either toadfish or sea robin. Toadfish
444
R. W. ROOT
blood i- biilTered the least of all.
between it and sea robin blood.
However, there is little difference
DISCUSSION
In the work on the effect of carbon dioxide on the oxygen capacity
and on the oxygen dissociation curves a suggestive series of results
were obtained. In mammalian hemoglobin the usual effect of carbon
dioxide is purely on the oxygen dissociation constant, a simple Bohr
effect with no upset in the original oxygen capacity of the particular
hemoglobin studied. The hemoglobins of these fishes, however, seem
to be affected by carbon dioxide in a manner more complicated. The
data suggest that some of the oxygen-binding groups of the hemoglobin
601
55
JO
1 K,. ''. Carbon dioxide-absorption curves for m.H -ki-rrl Mood at 20° C. The
dots are for reduced, and the circles for oxygenated blood.
molecule have become inactive. If the hemoglobin molecule combines
with four molecule^ of oxygen, as has been suggested by Adair (1925), it
\\ould appeal as if carbon dioxide were inactivating one or more of the
tour prosthetic groups involved in binding oxygen. In other words, it
RESPIRATORY FUNCTION OF BLOOD OF MARINE FISHES 445
would look as though the hemoglobin-oxygen reaction were stopping
off at one or more of the intermediate compound stages, depending
upon how much carbon dioxide is present, instead of the reaction being
carried completely through the four theoretical steps presented by
Adair.
To illustrate this point attention is recalled to the results on the
direct effect of carbon dioxide on oxygen capacity. In the case of
toadfish blood there is produced a maximum loss of about 55 per cent
30
I'
\»
OX>
7.00 7.2S 7.50 T7f SOO
6.30 tits 1.00 ns rx> 775 aoo
/>*•
FIG. 10. FIG. 11.
FIG. 10. BHCO3 : pH curves for toadfish blood at 20° C. The dots are for
reduced, and the circles for oxygenated blood.
FIG. 11. BHCO3 : pH curves for sea robin blood at 20° C. The dots are for
reduced, and the circles for oxygenated blood.
in oxygen capacity in the presence of carbon dioxide, and in sea robin
blood about 70 per cent. Mackerel blood under the same conditions
experiences a loss of slightly over 60 per cent. Interpreting this situa-
tion on the basis of inactivation of oxygen-binding groups, toadfish
hemoglobin has two of the four groups inactivated. Thus, allowing for
experimental errors, the oxygen capacity drops to a point approxim-
ately 50 per cent lower than the original figure for oxygen capacity ob-
tained when the blood was equilibrated in air. Sea robin hemoglobin,
and perhaps mackerel, has three of the four groups inactivated. Thus
the new figure for oxygen capacity obtained in the presence of consider-
able carbon dioxide is approximately 75 per cent lower than the
original. As has been pointed out previously, these marked drops in
oxygen capacity occur at definite ranges of pH.
It will be recalled that reference was made to the peculiar tendency
of the oxygen dissociation curves (most marked in the case of those for
the toadfish and sea robin) to appear to reach a limit considerably be-
fore the 100 per cent oxygen-saturation point was reached. It seems
reasonable to suppose that the phenomenon of inactivation of oxygen-
binding groups affords an interpretation of this situation.
There is no doubt but that there is danger in carrying the foregoing
interpretation too far. The author wishes to emphasize the fact that
446
R. W. ROOT
the idea of inactivity brought forth in this paper is purely suggestive.
Data are far too few to warrant any definite conclusion. If the data
really mean that certain prosthetic groups are inactivated, then the
oxygen-dissociation curves should present asymptotic relationships
from the point of minimum oxygen tension at which the remaining
active groups are saturated with oxygen up through oxygen tensions
far above those used in these experiments. At the same time the same
marked loss in oxygen capacity in the presence of carbon dioxide should
be capable of demonstration even though the blood were equilibrated in
pure oxygen. It is regretted that higher oxygen tensions were not
SO
45
40
&
^ *
<v;
5,
?•
^ *
<$
ts
Jo
5
o
6
V
\
A
A
\
\
\
\
\
\
\
\
°\
\
\
V
\
4o
•s
k«-o
»
73 7.OO 7.25 ISO 7.15 6.OO S.2S
/'A
Fir,. ]2. I'.IICOs : pH curves for mackerel blood at 20° C. The dots are for
reduced, and the circles for oxygenated blood.
used, for it seems that if such had been the case the idea of inactivity
would have had either a stronger case in its favor or been thrown out
entirely. It may be that the entire situation is a greatly exaggerated
I '" 'In Hied , and all that is necessary is higher oxygen tensions to bring
back the original oxygen capacity.
If the idea presented in this paper proves upon further experimen-
tation to be correct, we have before us a mean- of furthering the study of
Adaii'V theory of the combination of oxygen with hemoglobin.
A-ide from the physical chemistry of fish hemoglobin, the relation of
the data pre-ented to the life of the fish is interesting. We find a cor-
relation betueen the transportation of oxygen and the environment and
RESPIRATORY FUNCTION OF BLOOD OF MARINE FISHES 447
habits of the fishes. The sluggish fishes have bloods of low oxygen
capacity, and the active of high capacity. Thus, there is evidence of
adjustment between oxygen capacities and oxygen requirements, for,
as Hall (1929) has shown, the sluggish fishes do not consume as much
oxygen per unit time as the active. Further evidence of adjustment is
shown in the form of the oxygen dissociation curves at low carbon
dioxide tensions. The toadfish hemoglobin, under these conditions,
becomes saturated with oxygen at a much lower tension than is the case
with the other two fishes. This may partially explain the ability of
this fish to live in water of abnormally low oxygen tension (Hall, 1930).
On the other hand, mackerel hemoglobin, in the presence of 1 mm. of
carbon dioxide, requires a considerably greater tension of oxygen to
become saturated than is the case for the other fishes studied. This
may account in part for the great susceptibility of the mackerel to
asphyxiation. Hall (1930) found that a mackerel requires a strong
circulation of oxygen-loaded sea water over its gills in. order to prevent
excessive oxygen- unsaturation of its blood, and consequent death due
to asphyxia.
The high sensitivity of all three hemoglobins to carbon dioxide indi-
cates that they are adjusted to an environment of low carbon dioxide
tension, such as the gills offer. Any one of the fishes examined would
experience considerable difficulty in getting sufficient oxygen were the
environment in which its gills are bathed loaded with free carbon
dioxide. Krogh and Leitch (1919) and Redfield et al. (1926) have
alluded to the apparent adjustment of the oxygen dissociation curves
to the environment and habits of animals. Krogh and Leitch offered
such a conclusion after working on the blood of fishes, while Redfield
and collaborators came to the same conclusion after investigating cer-
tain bloods containing hemocyanin. The work presented here corrobo-
rates their evidence.
With regard to the transportation of carbon dioxide by the blood of
marine fishes, this investigation shows that the amount bound by the
various bloods is not the same for all species. Directionally the same
differences occur as were found in the ability of the bloods to combine
with oxygen. Mackerel blood is not only able to bind greater quanti-
ties of oxygen, but is also able to bind greater quantities of carbon
dioxide than either toadfish or sea robin blood. This strongly suggests
that the greater concentration of hemoglobin in mackerel blood is re-
sponsible for the difference noted. It is known that hemoglobin affects
the height and slope of carbon dioxide-absorption curves. This has
been pointed out by Peters, Bulger, and Eisenman (1924) and others.
The writer, too, found that anaemic fish blood would not take up as
448
R. W. ROOT
murli (-.trlii m dioxide as normal blood of a species. It is generally rec-
o-m/ed that hemoglobin plays an important role in the transportation
<>t" i, ul. on dioxide. This has been shown by Van Slyke (1921) and
many other workers. However, just how close a relationship there is
between the hemoglobin concentration and the ability of fish blood to
carry carbon dioxide cannot be stated at this time.
The greater concentration of hemoglobin in mackerel blood may
alst • account for the fact that it is easier to demonstrate greater carbon
dioxide absorption by its reduced than by its oxygenated blood, than
to do it with either toad fish or sea robin blood.
The small amount of carbon dioxide found in the circulating blood
of these fishes is in agreement with the findings of Kokubo (1930) for
certain other marine species. At the same time the relatively poor
buffering ability of their blood agrees with data on other forms pre-
sented by Collip (1920), \Yastl 1 1928), and Kokubo (1930). The facts
that there is little carbon dioxide normally present in the blood of these
lishes, and that it is poorly buffered against carbon dioxide, again sug-
JOO
I 1C. 1 v Comparative oxygen dissociation curves. Cm\e 1 is for toadfish
Mood at 20° C. .iml 1 mm. carbon dioxide; curve 2 for human blood at 37.5° C. and
J(i Him. carliou dioxide; < m \ e 3 for turtle blood at 25° ( '. and 40 mm. carl. on dioxide;
curve 4 for carp blood al 1S° C. and 30 mm. carliou dioxide; and curve 5 for mackerel
blood at 20° C. and 1 mm. carbon dioxide tension.
t an adjustment of the bloods to sea water. There is a low carbon
dioxide ten-ion in the gill of a marine fish, a fact necessarily correlated
RESPIRATORY FUNCTION OF BLOOD OF MARINE FISHES 449
with the low carbon dioxide tension in sea water. At the same time,
because of low metabolic rate, a fish produces relatively small quanti-
ties of carbon dioxide. Mammalian blood must, by virtue of the high
alveolar carbon dioxide tension and the greater metabolic activity on
the part of the animal, be prepared to handle larger quantities of carbon
dioxide than the blood of a fish. The situation as it stands appears to
point to adjustment on the part of both fish and mammal blood to the
particular physiological, morphological, and ecological differences that
concern the two types of vertebrates.
00
9o too
Pcozmm.J{q
FIG. 14. Effect of carbon dioxide on the "unloading tension" (Po2 when blood
is half saturated) of various vertebrate bloods. Curve 1 is for sea robin blood at
20° C.; curve 2 for toadfish blood at 20° C.; curve 3 for mackerel blood at 20° C.;
curve 4 for human blood at 37.5° C.; curve 5 lor turtle blood at 25° C.; and curve 6
for carp blood at 18° C.
The calculated pH of fish blood is less than that of sea water. One
may wonder how the blood maintains a lower pH. The facts that the
blood is poorly buffered, and that it maintains a carbon dioxide tension
normally higher than that of sea water probably account for the lower
pH.
In comparing the data presented in this paper with similar data on
other vertebrates, several interesting differences are brought out. In
Fig. 13 a family of oxygen dissociation curves is shown. Conditions
30
450
R. W. ROOT
ha\e been chosen in such a manner as to make the curves fairly near
alike. The oxygen dissociation curve for human blood has been con-
structed from the data of Bock, Field, and Adair (1924); that for the
turtle from Southworth and Redfield's (1926) work; and that for the
carp from \Yastl's (1928) data. The most noticeable thing about these
curves is the diversity of conditions under which they were established.
The only \\ ay one can make them rocml >le each other fairly closely is to
establish them under widely different conditions of temperature and
carbon dioxide tension.
Fin. 15. ' i.iiive carbon dioxide-absorption curves for reduced blood
i ept for turtle). Curves 1 and _' arc for turtle blood at 25° C.; curve 3 for human
Mood at 15 °C; cur\c 1 for frog blood at 15° C.; curve 5 for mackerel blood at 20° C.;
curve o for carp blood at 18° C.; curve 7 for sea robin blood at 20° C.; and curve 8 for
io..dtisli blood at 20° C.
In order to show how these same bloods arc affected differently by
( arbon dioxide l;ig. 1 1 has been constructed. ( >nc can see at once that
i he effect of c,i! linn dioxide on marine fish 1)1 ood is profoundly different
from its effect on either human, turtle, or carp blood.
The forev;"iii- c ompai i^ons point out \\cll the specificity of hemo-
globin in nature that Harcroft (1928) stresses. The significance of
>pcciliciiy is great. \\Vre all hemoglobins alike many animals would
RESPIRATORY FUNCTION OF BLOOD OF MARINE FISHES 451
not be able to exist under the conditions of their environment, or of
their assumed structural and functional characteristics.
For the purpose of showing the differences between the carbon di-
oxide-absorption curves of various vertebrate bloods Fig. 15 is pre-
sented. The data plotted are for reduced blood, except in the case of
the turtle. The curves for human, frog, and carp bloods have been
constructed from the data of Wastl and Seliskar (1925), and Wastl
(1928) ; and those for the turtle from Southworth and Redfield's (1926)
s
s
3o
OX,
Jo)
®
t>.$0 b.75 700 7.23 ISO 77S 8.OO 025 &SO
FIG. 16. Comparative BHCO3 : pH curves for reduced blood (except for turtle) •
Curve 1 is for turtle blood at 25° C.; curve 2 for frog blood at 15° C.; curve 3 for hu-
man blood at 15° C.; curve 4 for carp blood at 18° C.; curve 5 for mackerel blood at
20° C.; and curve 6 for toadfish blood at 20° C.
data. The curves show that the blood of fishes is characterized by a
relatively weak, those of the frog and turtle by a relatively strong, and
that of the human by a more or less intermediate carbon dioxide-
combining power. Human blood yields the steepest carbon dioxide-
absorption, which means that it is buffered the best. These curves
have been plotted at as near the same temperature in all cases as pos-
sible, since it has been shown by Warburg (1922), Stadie and Martin
(1924), and Cullen, Keeler, and Robinson (1925) that temperature af-
fects the carbon dioxide-combining power of blood.
In order that the buffering ability of several vertebrate bloods might
be compared Figs. 16 and 17 were constructed. Data other than the
author's have been taken from the previously mentioned sources and
the pH or cH calculated on a basis comparable to the calculations made
for marine fish blood. In Fig. 16 the BHCO3 : pH relationships are
452
R. W. ROOT
shown; in I it;. 17 the 10~8 X cH : Pco2 relationships. In the first
figure the more nearly parallel the curve runs with respect to the
al'M-i— a the more poorly the blood is buffered. The results here indi-
cate that toadfish blood is the poorest buffered, while human blood is
tin- best buffered. There appears to be little difference in the other
bloods. In the second figure the steeper the curve is, the poorer the
blood is buffered against carbon dioxide. The results obtained here
indicate that toadfish and sea robin blood are relatively poorly buffered,
\\ hile frog, turtle, and human blood are relatively well buffered. Carp
and mackerel blood are more or less intermediate with respect to the
others, resembling, however, the bloods of the higher vertebrates
slightly more than those of the toadfish and sea robin.
z/
19
'<
15
.
<§>
•to sv t,i 70 eo 90 too jjo
FIG. 17. Comparative 10~8 X cH : Pco2 curves for reduced blood (except for
turtle). Curve 1 is for toadfish blood at 20° C.; curve 2 for sea robin blood at 20° C.;
curve 3 for carp blood at 18° C.; curve 4 for mackerel Mood ai 20° C.; curve 5 for
frog blood at 15° C.; curve 6 for human blood at 15° (".; and curve 7 for turtle blood
at 25° C.
There is another point of interest about fi^s. K>.md 17. Kegard-
ol tin- --lope of the curves, at any given pi I the bloods do not have
the -.inic l!l !('<)., miik-nt; likewise at any givrn cH they are not sub-
jected to the ^[me. carbon dioxide tension. This may be explained by
tact ih.it the carbon dioxide-absorption lexel is quite different for
RESPIRATORY FUNCTION OF BLOOD OF MARINE FISHES 453
the different bloods. The higher the level at a given carbon dioxide
tension the more the hydrogen ion concentration is displaced in the al-
kaline direction. Southworth and Redfield (1926) have shown that as
far as turtle blood is concerned the characteristically high level of the
carbon dioxide-absorption curve is due to high BHCO3 in the plasma
and the relatively small amount of hemoglobin present to act as an
acid in dissociating carbon dioxide from its salt. Perhaps the same
thing holds true for frog blood. It is interesting to note that in the case
of toadfish and sea robin blood the dissociation of carbon dioxide is
quite complete even though there is a low hemoglobin concentration.
The differential buffering ability of the bloods may possibly be
explained on the basis of the nature of the adjustments that vertebrates
have undergone in going from an aquatic to a terrestial environment.
The acquirement of lungs and a higher rate of metabolism has made
necessary a greater buffering defense.
SUMMARY
1. The oxygen capacities of marine fish bloods are quite different
for different species. The greatest difference is between the typically
sluggish and active forms, the former having bloods of low, and the
latter bloods of high oxygen capacity. There is a general correlation
between oxygen capacity and corpuscle count, corpuscle volume, and
iron content.
2. Studies on the oxygen dissociation curves of marine fish hemo-
globin, and on the effect of carbon dioxide on the oxygen capacity have
brought forth the suggestion that the effect of carbon dioxide on the
hemoglobins of these fishes is not solely on their oxygen dissociation
constants, but that there is an inactivation of certain of the prosthetic
groups concerned in binding oxygen in the hemoglobin molecule, caus-
ing a marked decrease in oxygen-combining power of the bloods.
The most marked evidence of inactivation occurs at definite ranges of
carbon dioxide tension and pH for the different bloods.
3. The carbon dioxide-combining power of fish bloods appears to be
correlated with hemoglobin concentration. Mackerel blood with high
hemoglobin absorbs more carbon dioxide than toadfish blood, which
has a low hemoglobin concentration.
4. Reduced fish blood will absorb slightly more carbon dioxide than
oxygenated blood. For sea robin and toadfish bloods the range of
carbon dioxide tension where this can be demonstrated is short, being
between about 2 and 25 mm., while it is longer for mackerel blood,
being about 2 to 95 mm.
5. There is a differential buffering ability shown by these bloods,
mackerel blood being buffered the best and toadfish the poorest.
454 R. W. ROOT
6. Comparative studies of vertebrate bloods strengthen the idea of
-prciiidty of hemoglobins. Those of the marine fishes are far more
-riir-itivc to carbon dioxide than those of the carp, turtle, and human.
7. Comparative studies on carbon dioxide transportation show that
turtle and frog bloods have a relatively great, fishes a relatively small,
and human blood a more or less intermediate carbon dioxide-combining
power. The bloods also vary considerably in their buffering capacity,
human blood having the greatest and toadfish blood the least.
8. The general results of this investigation point to an adjustment
on the part of the blood of marine fishes to a sea-water environment,
and the habits or characteristics of the fishes. At the same time the
comparative studies indicate marked differences between the bloods of
fishes and terrestrial vertebrates. These differences can perhaps be ac-
counted for on the basis of the new morphological and physiological
features that terrestrial vertebrates have acquired, along with change
in en\ ironment, which have made necessary correlative changes in the
respiratory function of the blood.
I wish to express to 1 )r. F. G. Hall my profound appreciation for the
many timely suggestions and criticisms that he offered during the
progress of this work. I wish to thank various members of the Duke
I ni\a>ity /oology Department, and of the United States Bureau of
I isheries, particularly Dr. I. 1C. Gray, Dr. O. E. Sette, Dr. A. S. I'earse,
and Mr. S. l\. Tiptoii. I also wish to thank Dr. A. C. Red field of
Harvard I 'niviTMty for the many helpful suggestions that he has
given me.
LITI.K MURE CITED
ADAIR, G. S., 1925. The I Icmoylobin System. VI. The oxygen dissociation curve
of hemoglobin. Jour. Biol. Client., 63: 539.
AUSTIN, J. II., i .. I.. CULLEN, A. I'.. HASTINGS, 1 . C. \b I i \\, J. 1'. I'KIKUS, \\i.
D. D. VAN SLYKE, 1922. Studies of Gas and Elenn.Kie Equilibria in
Blood. I. Technique for collection and analysis of blood, and for its
saturation with gas mixtures of kno\\n < •omposit ion. Jour, ttiol. Client.,
54: 121.
BARCROFT, J., 1928. The Respiratory Function of the I'.lood. Part II. Hemo-
globin. < ambridge.
\. V., II. I ii I.D, AND G. S. ADAIR, 1924. The <)\\-en and Carbon Dioxide
Absorption Curves of Human Blood. Jmir. Hi-nl. Client., 59: 353.
BOHR, ( '., 1905. Absorptions cocTficienten des Blutes und des Blutplasmas fur Gase.
Sk,ni,L Anh.f. Physiol., 17: 104.
CHRISTIANSEN, J.. C. G. DOUGLAS, AND J. S. HALDANE, I'M 4. The Absorption and
Dissociation of ( arbon Dioxide by Human I Hood. Jour. Physiol., 48: 244.
< • .1.1. IP, J. 15., 1920. The Alkali Reserve of Mai me I i>h and Invertebrates. The
excretion "t Iaibon dioxide. Jour. Bin!. Client., 44: 329.
< i ii i \. i ,. I... II. K. Ki I.I.KK, AND II. \V. ROBINSON, 1925. The pK' of the Hender-
son I lasselbalrh Equation for llydrion Concentration of Serum. Jour.
Biol. Cliem., 66: 301.
RESPIRATORY FUNCTION OF BLOOD OF MARINE FISHES 455
GAARDEK, T., 1918. Uber den Einfluss des Sauerstoffdruckes auf den Stoffwechsel.
II. Nach Versuchen an Karpfen. Biochem. Zeitschr., 89: 94.
GRAY, I. E., AND F. G. HALL, 1930. Blood Sugar and Activity in Fishes with Notes
on the Action of Insulin. Biol. Bull., 58: 217.
HALL, F. G., I. E. GRAY, AND S. I.KPKOYSKY, 1926. The Influence of Asphyxiation
on the Blood Constituents of Marine Fishes. Jour. Biol. Chem., 67: 549.
HALL, F. G., AND I. E. GRAY, 1929. The Hemoglobin Concentration of the Blood of
Marine Fishes. Jour. Biol. Chem., 81: 589.
HALL, F. G., 1928. Blood Concentration in Marine Fishes. Jour. Biol. Chem., 76:
623.
HALL, F. G., 1929. The Influence of Varying Oxygen Tensions upon the Rate of
Oxygen Consumption in Marine Fishes. Am. Jour. Physiol., 88: 212.
HALL, F. G., 1930. The Ability of the Common Mackerel and Certain Other Marine
Fishes to Remove Dissolved Oxygen from Sea Water. Am. Jour. Physiol.,
93:417.
HASSELBALCH, K. A., 1917. Die Berechnung der Wasserstoffzahl des Blutes aus der
freien und gebundenen Kohlensaure desselben, und die Sauerstoffbindung
des Blutes als Funktion der Wasserstoffzahl. Biochem. Zeitschr., 78: 112.
HASTINGS, A. B., AND J. SENDROY, JR., 1925. The Effect of Variation in Ionic
Strength on the Apparent First and Second Dissociation Constants of
Carbonic Acid. Jour. Biol. Chem., 65: 445.
HASTINGS, A. B., J. SENDROY, JR., AND D. D. VAN SLYKE, 1928. Studies of Gas and
Electrolyte Equilibria in Blood. XII. The value of pK' in the Hender-
son-Hasselbalch equation for blood serum. Jour. Biol. Chem., 79: 183.
HENDERSON, L. J., 1908. The Theory of Neutrality Regulation in the Animal
Organism. Am. Jour. Physiol., 21: 427.
HENDERSON, T., 1918. Applications of Gas Analysis. IV. The Haldane gas
analyzer. Jour. Biol. Chem., 33: 31.
JOBES, F. W., AND M. E. JEWELL, 1927. Studies on the Alkali Reserve of the Blood
of Ameriurus nebulosus from Acid and Basic Waters. Trans. Am. Micros.
Soc., 46: 175.
JOLYET, F., AND P. REGNARD, 1877. Recherches physiologiques sur la respiration des
animaux aquatiques. 1 and 11. Arch. Physiol. Norm, et Path., 4: 44, 584.
KAWAMOTO, N., 1929. Physiological Studies on the Eel. I. The seasonal variation
of the blood constituents. II. The influence of temperature and of the
relative volume of the red corpuscles and plasma upon the haemoglobin
dissociation curve. Sci. Rep. Tohoku Imper. Univ., Series 4, 4: 635.
KOKUBO, S., 1927. Contribution to the Research on the Respiration of Fishes. I.
On the hydrogen ion concentration and the CO2-gas content and capacity of
fish blood. Sci. Rep. Tohoku Imper. Univ., Series 4, 2: 325.
KOKUBO, S., 1930. Contribution to the Research on the Respiration of Fishes. II.
Studies on the acidosis of fishes. Sci. Rep. Tohoku Imper. Univ., Series 4,
5: 249.
KROGH, A., AND I. LEITCH, 1919. The Respiratory Function of the Blood in Fishes.
Jour. Physiol., 52: 288.
NICLOUX, M., 1923. Action de 1'oxyde de carbone sur les poissons et capacite respira-
toire du sang de ces animaux. Compt. r. Soc. Biol., 89: 1328.
PETERS, J. P., H. A. BULGER, AND A. J. EISENMAN, 1923. Studies of the Carbon
Dioxide Absorption Curve of Human Blood. I. The apparent variations
of pKi in the Henderson-Hasselbalch equation. Jour. Biol. Chem., 55: 687.
PETERS, J. P., H. A. BULGER, AND A. J. EISENMAN, 1924. Studies of the Carbon
Dioxide Absorption Curve of Human Blood. IV. The relation of the
haemoglobin content of the blood to the form of the carbon dioxide-absorp-
tion curve. Jour. Biol. Chem., 58: 747.
TOWERS, E. B., 1922. The Alkaline Reserve of the Blood of Fish in Relation to the
Environment. Am. Jour. Physiol., 61: 380.
456 R. W. ROOT
REI>I IKI.D, A. ('., T. COOLIDGE, AND A. L. HURD, 1926. The Transport of Oxygen
ami Carbon Dioxide By Some Bloods Containing Hemocyanin. Jour. Biol.
>!., 69: 475.
Ki.:'i II-:M>. A. ( '.., AND E. D. MASON, 1928. The Combination of Oxygen and Hydro-
chloric Acid with the Hemocyanin of Limulus. Am. Jour. Physiol., 85: 401.
Sot i HWORTH, F. C., AND A. C. REDFIELD, 1926. The Transport of C>as by the Blood
of the Turtle. Jour. Gen. Physiol. , 9: 387.
Si \I»IE, \Y. C., AND K. A. MARTIN, 1924. The Thermodynamic Relations of the
Oxygen- and Base-combining Properties of Blood. Jour. Biol. Cheni., 60:
191.
Si Mm., \Y. C., AND K. R. HAWES, 1928. Studies on the Oxygen-, Acid-, and Base-
combining Properties of Blood. IY. The apparent first dissociation con-
stant, pK/, of carbonic acid and the activity coefficient of the bicarbonate
ion in solutions of hemoglobin, methemoglobin, cyanhemoglobin, and nitric
oxide hemoglobin at varying ionic strengths. Jour. Biol. Client., 77: 2<>5.
STADIE, YV. ('., 1('2S. Studies on the Oxygen-, Acid-, and Base-combining Properties
of Blood. Y. Extension of the Debye-Hiickel theory of ionic interaction
to hemoglobin, bicarbonate-sodium chloride systems. Jour. Biol. Chetn.,
77: 303.
TRENDELENBURG, P., 1912. t'ber die Sauerstofftension im Blute von Seefischen.
Zeitschr.f. Biol., 57: 4<>5.
\'AN SLYKE, 1 ). I )., 1921. The Carbon Dioxide Carriers of the Blood. Physiol. AY;'.,
1: 141.
VAN SI.YKE, D. D., H. Wu, AND F. C. Mel. BAN, 1923. Studies of Gas and Electrolyte
Equilibria in the Blood. Y. Factors controlling the electrolyte and water
distribution in the blond. Join. Biol. Cliciti., 56: 765.
VAN SI.YKK, I). I)., AND J. M. NKILL, 1924. The Determination of Gases in Blood
and Other Solutions by Yacuum Extraction and Manometric Measurement.
I. Jour. Biol. Clif»i., 61: 523.
YAN SI\KI., D. D.. A. I'). HASTINGS, C. D. Mi RRAY, AND J. SENDROY, JR., 1925.
Studies of Gas and Electrolyte Equilibria in Blood. VIII. The distribu-
tion of hydrogen, chloride, and bicarbonate ions in oxygenated and reduced
blood. Jour. Hint. Client., 65: 701.
YAN SLYKE, D. D., AND J. SENDKOY, JK., 1927. Carbon Dioxide Factors for the
Manometric Blood Gas APIMIMI us. Jour. Biol. Chew., 73: 127.
WARBURG, E. J., 1922. Studies on Carbonic Acid Compounds and Hydrogen Ion
Activities in Blood and Sail Solutions. A contribution to the theory of the
equation of L. J. Henderson and K. A. I lasselbalch. Biochem. Jour.,
16: 153.
WASTL, H., AND A. SELISKAR, 1925. Observations on the Combination of Carbon
Dioxide in the Blood of the Bull Frog (Rana cato . Jou-r. Physiol.,
60: 264.
WASTL, II., 1928. Beobachtungen iiber die Blutgase des Karpfenblutes. Biochem.
Zeitschr., 197: 363.
THE RESPIRATION OF PUFFER FISH
F. G. HALL
(From the Department of Zoology, Duke University, and the U. S. Bureau of
Fisheries, Woods Hole, Mass.)
The mechanism of respiration of fishes which live in the sea offers
an attractive and productive subject for study. The ocean is stable and
uniform and therefore a favorable environment for living organisms.
An abundant supply of oxygen is usually present. The hydrogen ion
concentration varies only in a range which is close to the optimum for
physiological processes, especially for the elimination of carbon dioxide.
The temperature of the ocean as compared with freshwater and land
conditions is relatively uniform. Moreover, sea water is similar in
constitution to the internal fluids of marine organisms. Since most
vertebrates have mechanisms for maintaining conditions within their
bodies more or less constant and since fishes are the last of typically
marine vertebrates to evolve, it seems important to study the factors
which vary in sea water and which in some manner influence the respira-
tory exchange of gases between fishes and their surroundings.
Fishes breathe dissolved gases from water which they pump over
their gills. The mechanism for external respiration consists in most
fishes of rhythmical suction of water into the oral cavity and its subse-
quent expulsion through the gill clefts. During inspiration the mouth
is opened and the oral cavity enlarged by the lateral expansion of its
walls. When the oral cavity is closed the expiratory process begins.
By the lateral contraction of the oral walls water is driven through the
gill clefts and over the gill filaments. The branchial arches are spread
apart during the expiratory phase, thus permitting all of the filaments
to come into direct contact with the circulating water. The gas ex-
change between the blood and water takes place through the walls of the
filaments.
Considering the general mechanics of external respiration as shown
by fishes, several problems come to mind. How much water is pumped
in a single respiratory cycle? How much of the dissolved oxygen is
removed from sea water as it passes the gills? When an increased oxy-
gen supply is required, which plays the more important role — an increase
in the ability of the gills to absorb oxygen from the sea water; an in-
crease in the volume of water pumped by a single respiratory cycle; or
457
458 F. G. HALL
an iiu n tin- number of respiratory cycles per unit time? Consid-
eration is given to each of these possibilities in the following pages.
Three physico-chemical factors which may vary in tin- external me-
dium and affect the equilibrium which the organism maintains in its
internal environment are temperature, oxygen tension, hydrogen ion
concentration (carbon dioxide tension and hydrogen ion concentration
/vr sc). By varying the:-e [actors in the investigations to be described,
a means of studying certain phases of the general problem of respiration
was found.
The most extensj\e studies bearing on the problems of fish respira-
tion are those of Winterstein (1908). He used the fresh-water fish,
Lriicisciis erythrophtalmus. Fishes under observation were held fast
by a clamp, while a constant stream of water of a known oxygen tension
was passed over the gills by means of a thick canula fastened in the
mouth of the fish. The amount of oxygen used up was determined.
This is perhaps the- simplest and most direct method that has been de-
vised tor the- determination of the respiratorv exchange in fishes. 1 low-
ever, as Wintcrstein has pointed out. one must keep in mind that the
fishes are breathing somewhat abnormally. \Yhen fishes have water
forced over their gills, they may not respire in the same way as if
they were pumping the- water over the gills in the natural manner. He
concludes from his experiments that the' oxygen consumption is inde-
pendent of oxvgen tension of the surroundings within wide limits of
magnitude, and that the ntili/ation of oxygen is in inverse proportion to
the (lowing velocity. Jlenxe (1910) has also shown that oxygen con-
sumption in certain fishes is not influenced to any great extent by the
oxygen tension of the surrounding water. llis results are expressed in
arbitrary values and are not particularly constant.
Gaarder ( 1'MX) has performed an interesting experiment on the
fresh water car]), llis paper is stimulating and thoughtful. However,
it discusses onlv a few analyses and has the disadvantage that the gills
were subjected to forced ventilation and therefore were perhaps not
fiinetioniiig naturally. ( iaarder had the misfortune, it appears, of be-
ing quoted inaccurately, being said to conclude that oxygen consumption
is within wide limits proportional to oxygen tension. Another author
quotes him as believing that oxygen consumption is independent of oxy-
gen ten-ion. The writer understands ( iaarder's conclusion to be that
consumption is uninJlueiieed so long as the hemoglobin of the blood is
not fully saturated; when oxygen and the oxygen tension of the physi-
cally dissolved oxygen is raised considerably, then oxygen consumption
shows an increase-.
Powers (1922, 1929) and Powers and Shipe (1928) have shown
RESPIRATION OF PUFFER FISH
459
that carbon dioxide tension and pll have a pronounced effect on the
respiration of fishes. Powers (1930) has given an excellent summary
of the relation between pH and aquatic animals.
FIG. 1. Apparatus used for the determination of the influence of environmental
factors on the respiration of puffer fishes.
METHODS
The puffer fish, Spheroidcs macitlatits (Bloch and Schneider), was
used in the writer's investigations because it was particularly adapted to
such a study. The rounded shape of the opercular aperture, which is
considerably reduced in size as compared with other fishes, makes this
species especially advantageous. Glass tubes may be inserted through
460 F. G. HALL
the opercular openings \vithout apparent injury. All of the water
pumped by the fish lor respiratory purpose will then flow through the
iss tubes and samples can be collected for analyses. Fishes carrying
such tubes will lie quietly for hours apparently breathing normally, and
will live, in this condition for several weeks. Tubes were inserted in
the opercular openings of puffers about three or four days previous to
u^ing them for the respiration experiments. Thus the animals became
accustomed to breathing in such a manner.
The apparatus used in these experiments is shown in Fig. 1. The
fish was submerged in a chamber (c), which had a capacity of 4 liters.
A reservoir (7^) to which flowing water was admitted and which also
contained a funnel out of which all of the excess water flowed was
connected to the chamber by a hole one inch in diameter. Two side
compartments (s) were so arranged that they could be connected with
the glass tubes inserted in the opercular opening of the fish. A funnel
was placed in each side' compartment at the same level as that in the
reservoir (/?). The height of the funnels in each case was adjusted so
that before the fish was connected to the side compartments water en-
tering the reservoir would flow out through the funnel in the reservoir
but would not flow out through the funnels in the side compartments.
Thus only a very slight exertion on the- part of the fish was required to
pump water from the chamber in which it was submerged to the fun-
nek in the side compartments. Irishes were placed in the chamber so
that their months wen- close to the hole leading from the reservoir.
Thus a fresh flowing supply of water was always available. It was
not found necessary to either ana-si lu-ti/e these fishes or to clam]) them.
If they were left undisturbed by outside factors they would remain
quiet for hours.
The quantity of water pumped per minute.- was measured by use of
volumetric flasks placed under the funnels, and a stopwatch. The quan-
tity of water pumped through the right and left gill chambers was taken
separately. Analyses of tin- dissolved ox\gen was made on the water
before it entered the fish's mouth and after it bad been pumped into the
side compartments. The well known \Yinkler method as modified by
Birgc and Jnday was employed. Care was taken not to expose the
water to air in taking the samples.
In experiments where the influence of temperature was studied,
water was cooled to the desired temperature by passing through coils
in a constant temperature bath. A range of 10° C. was used since
pullers do not readily adjust themselves to a lower temperature than
Ki -11° t . or higher than 23°-24° C. The temperature range chosen
for this experiment was 12°-22° C.
RESPIRATION OF PUFFER FISH
461
The hydrogen ion concentration of the water was measured by
colorimetric means. Consequently the analyses do not represent a pre-
cise measurement or an absolute value since salt errors are introduced.
No corrections have been made for salt errors. The pH determinations
must be taken only as of relative values. In one type of experiment
the hydrogen ion concentration of the water was controlled by the addi-
tion of carbon dioxide gas. In a second type- hydrochloric acid was
FIG. 2. Graph showing the influence of temperature on the respiration of
puffer fishes. Respiratory rhythm (R) in respirations per minute; oxygen con-
sumption (M) in cc. of oxygen per kilogram per hour; water pumped through
branchial cavity (W ) in deciliters per hour; percentage of dissolved oxygen (0)
removed from the affluent water.
added to the water and the carbon dioxide formed was driven off by
aeration.
Sea water of different oxygen tensions was procured by boiling and
subsequent mixing with normal sea water. In this manner sea water of
any desired oxygen tension could be obtained.
462
F. G. HALL
RESULTS
The results of the first experiment are graphically indicated in
I. They slum- the influence of temperature on the respiration of
puffer lilies. Ten individuals were submitted to various temperatures
indicated <>n the graph and the results for each were averaged. It
may IK- observed that the respiratory rhythm (R), rate of metabolism
!.W). and quantity of water pumped per minute (W) increased pro-
gressively with increase in the temperature of the surrounding water.
100
80
60
20
0
R
M
W
0
8.5
6.0
6.5
6.0
7.5 7.0
ph
I;K,. 3. Graph showing the influence of pll with \»\\ ( '< >.• tensions on respira-
tion of pnffiT fislu-s. Scale and legend as in IM.U. -.
The percentage of the dissolved oxygen absorbed from the surrounding
water, however, did not increase appreciably. At 20° C, which was
approximately the temperature of sea water in the \Yoods Hole Region
at the time these experiments were conducted, puffer fishes had an
average rhythm of 80 respirations per minute, pumped 6 liters of water
over their -ills in an hour, absorbed 45 per cent of the dissolved oxygen
from the water, and consumed on the average (>_! cc. of oxygen per
kilogram of body weight in an hour.
INSPIRATION OF PUFFER FISH
463
The second experiment shows the- effects on the respiration of puffer
fishes of varying the hydrogen ion concentration of the surrounding
water by the addition of hydrochloric acid to sea water (and subsequent
aeration in order to remove excess carbon dioxide). The results ob-
FIG. 4. Graph showing the influence of pH with high COa tensions on respira-
tion of puffer fishes. Scale and legend as in Fig. 2.
tained with six individuals are averaged and summarized graphically in
Fig. 3. They show that decreasing pH per sc apparently inhibits the
rate of metabolism (^/), and the amount of water pumped by fishes.
464
F. G. HALL
The respiratory rhythm (R) is affected slightly. The percentage of
-in absorbed (O) decreases with increasing acidity.
The third experiment was devised to show how dissolving carbon
dioxide1 would affect respiration as compared with the effect of pH pro-
ducrd by hydrochloric acid in the previous experiment. Figure 4 repre-
its the average results obtained with puffer fishes. These indicate
that variations in carbon dioxide concentration expressed in terms of
the pH of the water which contains it have a much greater influence
on respiration of fishes than variations in pi I due to other factors.
The quantity of water pumped (IV) was markedly accelerated when
the sea water approached the acid side of neutrality. The rate of
metabolism was greatly inhibited. The- respiratory rhythm decreased
in rate accordingly. Fishes died when the pi 1 was lowered below 6.5,
while in the previous experiment no difficulty was experienced in sub-
mitting individuals to a pH of 6.0.
TABLE I
The percentage oj dissolved oxygen absorbed by puffer fish from sea water
of varying oxygen tensions at 20° C.
Dissolved Oxygen in cc. Per Liter
Percentage of Dissolved
Oxygen Absorbed
Affluent Wat.-r
Eflluent Water
4.68
2.16
46
4.00
1.84
46
3.10
1.49
48
2.31
1.10
47
1.14
0.58
45
0.98
0.45
46
The purpose of the fourth experiment was to determine the per-
centage of oxygen which fishes absorbed at different oxygen tensions.
These determinations were made on eight individuals at a constant
temperature of 20° C. Dissolved oxygen analyses were made on the
affluent water which was being sucked into tin- mouth of the fishes and
on the effluent water which was flowing out of the opercular opening
after it had passed the gills. The results obtained with each individual
were averaged and are shown in Table I. They indicate that fishes are
aMc to absorb from 45 to 48 per cent of the dissolved oxygen from sea
water regardless of wide variation in the tension of the dissolved oxy-
n in the affluent water.
RESPIRATION OF PUFFKR FISH 465
DISCUSSION
It is evident from the foregoing experiments that the puffer fish is
able to pump considerable water over the gills. The quantity of water
circulated through the gill clefts and over the gill filaments varies under
different circumstances (Fig. 2). When the temperature of the in-
spired water is increased, the oxygen consumption increases progres-
sively. Concomittantly more water is pumped by the fish. However,
the quantity of water which is pumped by a single inspiration and ex-
piration varies but little and remains relatively constant through a wide
range of temperature changes. The rate of respirations per minute, on
the other hand, shows a parallel increase with that of water pumped and
oxygen consumed by the organism. Similarly, the quantity of dis-
solved oxygen removed does not seem related or influenced by the
oxygen consumed, but remains at a fairly constant level. Between
temperatures of 12° and 22° C. the variation in the percentage of oxy-
gen removed from the inspired water was between 44 and 45 per cent.
It seems, therefore, that the need for an increased quantity of oxygen
with increasing temperature is obtained mainly by regulation of the
respiratory rhythm and not by the quantity of water pumped on each
inspiration or the quantity of oxygen removed from the inspired water.
The gills are apparently a very efficient mechanism through which
oxygen is absorbed into the blood. Gill filaments are made of numerous
lamella?, thereby increasing the absorptive surface. Capillaries supply
the lamellae with blood which passes into the general circulation. The
outer membrane of the gill filaments is very thin, only a few microns in
thickness. Through this membrane dissolved molecular oxygen passes
from the sea water into the blood and is there bound by hemoglobin. A
small quantity of molecular oxygen will also be found in the blood in
the same state as in sea water, i.e., physically dissolved. The oxygen
capacity of the blood of puffer fishes has been found to range from 8 to
10 volumes per cent.
When water is pumped into the mouth of the puffer, it is forced out
between the branchial arches in such a way that a great proportion of it
comes into contact with the gill lamellae. The gills are flattened and
elongated and are fairly close together when water is forced past them.
Thus their anatomical arrangement is particularly advantageous. Sev-
eral factors are to be considered in properly interpreting their function.
When a stream of water passes through a branchial cleft its velocity will
be greatest in the middle of the stream and least nearest the lamella!:.
Relatively more oxygen will consequently be absorbed from the water
nearer the lamellae than from that further away. If oxygen is to be
31
466 F. G. HALL
ab- from the water moving at the higher velocity it must diffuse
ther ra]>i<lly. The rate of diffusion will depend upon the pressure
gradient.
Thus the efficiency of the respiratory mechanism may in a way be
•••mimed by comparison of the gas tensions of the affluent and effluent
water. Figure 1 shows that at 20° C. puffers pump an average of 6
liters of water per hour over their gills, and that 45 per cent of the clis-
s> lived oxygen was removed from affluent water. This indicates a very
effective aeration of the gills. Such a conclusion is further snbstanti-
•1 by the fourth experiment, in which the oxygen tension of the
affluent water was changed through a series of tensions ranging from
0.9 cc. per liter to -l.X cc. per liter. It was found that about the same
percentage of oxygen was removed regardless of the oxygen tension.
The percentage- varied only from 45 to 48 per cent. This indicates that
the respiratory mechanism of gill aeration is equally efficient over quite
a wide range of oxygen tensions.
An interesting point which must be considered in investigations
concerned with the respiration of fishes is the absence of any mechani-
cal buffering means such as is present in the alveolar air of air-breathing
animals. Mammals particularly have a residual air supply which main-
lains a fairly constant CO2 and O2 tension so that moderate irregu-
larities in breathing only slightly change the gas ten-ions of the alveolar
air. Fishes, however, have their gills directly exposed to water and
have nothing comparable to alveolar air teutons. Their gills are di-
rectly exposed to the gas tension of the water in which they live. They
have apparently no means by which the L;.HS tensions to which their gills
are subjected may be altered. Since the amount of CO2 in sea water is
low, the CO., tension of the' water surrounding the gill filaments would
be much lower than the CO., tension in the alveolar air of lung-breathing
vertebrates. Investigations are now being conducted to determine the
CO2 tension of fishes blood and its role in the respiratory function of
the blood.
SUMMARY
1. A method is described for studying environmental factors which
affect the respiration of fishes.
2. An increase' in temperature of water surrounding puffer fishes is
followed by increased oxygen consumption by the fishes, a greater quan-
titv of water pumped through the branchial chamber, and a faster re-
spiratory rhythm. The percentage of dissolved oxygen absorbed re-
mains constant at all temperatures observed.
3. Increase in hydrogen ion concentration inhibits oxygen consump-
RESPIRATION OF PUFFER FISH 467
tion by marine fishes, but addition of CO2 has a more pronounced effect
than addition of HC1 at the same pi I.
4. The results indicate that marine fishes apparently remove dis-
solved oxygen from sea water by an efficient mechanism of gill aeration.
Fishes absorbed about 46 per cent of dissolved oxygen from sea water
at all observed oxygen tensions.
BIBLIOGRAPHY
GAARDER, T., 1918. B'wchcm. Zcltschr., 89: 94.
HENZE, M., 1910. Biochcm. Zcltschr., 26: 255.
POWERS, E. B., 1922. Jour. Gen. Physio!., 4: 305.
POWERS, E. B., 1929. Ecology, 10: 97.
POWERS, E. B., 1930. Am. Nat., 64: 342.
POWERS, E. B., AND L. M. SHIPE, 1928. Pub. Pugct Sound Biol. Sla., 5: 365.
WIXTERSTEIN, H., 1908. Pflitger's Arch., 125: 73.
Till-: RATE OF OXYGEN CONSUMPTION OF ASTER I AS
EGGS BEFORE AM) AFTER FERTILIZATION'1
PEI-SUNG TAX',
(From the Marine Binln^ical- Laboratory, ]\'ooils llolr. Muss.)
I
Since the account of Loeb and \\"astencys (1912) nineteen years
ago, no data have been made available on the rate of oxygen consump-
tion of Asteruis eggs before and after fertilization. In view of the
importance of such studies for the understanding of the mechanism of
development as well as that of cellular oxidation, it was considered
desirable to reinvestigate the subject, using the microrespirometer
technic. This method has the advantage over the Winkler method,
which Loeb and \Yasteneys used, in that slight changes in rate of
oxygen consumption can be detected at rather short consecutive time
intervals.
II
The microrespirometers employed were those described by \Yarbur-
1()J(>i. ( 'onical vessels of about three cubic centimeters capacity with
-ide arms and cylindrical insets for alkali were used. Half a cubic
< entimeter of egg suspension wa- placed in each vessel. In the experi-
ment- with fertilixed eggs, 0.1 CC. of sperm suspension was introduced
rither directly into the chamber containing the eggs or into the side-
arm to be mixed with the egg^ after a number of readings on the un-
fertili/ed eggs had been taken. The experiment \\ere conducted at
JS.o' C. and the manometers shaken al the rate ol 70 complete oscilla-
tions a minute \\ith an amplitude of 15 cm., which wa- demonstrated
to be adequate to insure the requisite mixing.
Eggs from single animals were used. The gonads \\ere removed
from the animal- with a pair of force])- after partially detaching the
appendages, and placed in about twenty-ii\e cul>ic centimeters ol sea
water. After the egg- were shed, they \\ere hllered through cheese
cloih into a 100 cc. beaker tilled with sea \\ater and concentrated by
decanting the supernatant liquid. A portion of the eggs was examined
about twenty minute- after removal for maturation, and only those
lot- "I ' \\ith 50 per cent or more mat in ation were- used in the ex-
I in part by a grant from the KockclVlln I < >m illation to the University
468
OS-CONSUMPTION OF ASTERIAS EGGS
469
periments. At the end of an experiment, the eggs were examined again
and the percentages of maturation or cleavage were recorded. The
experiments were conducted during July and August at a time past the
height of the breeding season and the number of satisfactory experi-
ments available for analysis was relatively few. Howrever, all experi-
ments showed good agreement qualitatively, and only the typical ones
are given here.
Ill
A series of experiments was conducted in the following manner:
Half-cc. portions of egg suspension were placed in four vessels with the
sperm in the side arms. After a number of readings at 5-minute inter-
vals with the eggs unfertilized, the sperm in the side arms of three of
the vessels was mixed with the eggs, an operation requiring less than a
minute, and readings at 5-minute intervals were continued for 100
minutes. The data are plotted in Fig. 1. In these graphs the ordinate
03
'5.
(fi
rt
0)
tu
OO__OO on
B
o o oo
o-o-
GO O (J
0 .
10 20 30 40 50 60 70 80 90 100 110 120
Time in Minutes
FIG. 1.
represents the relative rates of oxygen consumption and the abscissae,
time in minutes. Lines B, C and D are obtained from experiments in
which the eggs \vere fertilized after the fourth reading : line A represents
the control in which the eggs and the sperm remained separate during
the experiment. The arrow points to the time of fertilization. The
results indicate that there is no change, either temporary or permanent,
in the rate of oxygen consumption during the first 100 minutes after
fertilization, and the scattering of the points is almost identical in the
fertilized and the unfertilized eggs. This scattering is due, presumably,
to errors in reading the small changes on the manometers. The result
confirms the findings of Loeb and Wasteneys, and is unlike the case of
Arbacia eggs (e.g., Tang, 1931). It may be remarked that the per-
470
PEI-SUNG TANG
cei of fertilization and cleavage were somewhat low (less than 50
per cent of all eggs), although samples from the same lot of eggs kept
in a Syracuse watch glass at room temperature (25° C.) showed as
much as 85 per cent cleavage. These low percentages would diminish
hut not mask the respiratory changes due to fertilization if they were
present.
21
18
•
E
E
<-> 15
—
a
E
en
O
U
c
a>
L2
15 45 75 105 l.vS 105
Time in Minutes
FIG. 2.
In a second series of experiments the eggs in two of the respirometer
vessels were fertilized immediately before the experiments \\ere started,
tuo other \csscls contained the same amount <>t" ei^s hut unfertilized,
.ind a fifth vessel contained a known amount of the sperm suspension
used in the hrst two vessels. The rates of oxygen consumption were
lollowed for three hours. When the experiments were performed in
this \\ ay, over 50 per cent of the eggs had cleaved to 8 and 16-cell stages
at the end of the experiment, and over 80 per cent of the eggs had
matured. The data are presented in 1 ii;. 2, in which the ordinate
represents the amount of oxygen consumption in cubic millimeters and
©.-CONSUMPTION OF ASTERIAS EGGS 471
the abscissa, the time in minutes after the closing of the manometers.
The values for the fertilized eggs with sperm are plotted as line A , those
for the unfertilized eggs, B; and those for the sperm, C. The broken
line D is the corrected curve for the fertilized eggs minus the sperm,
i.e., A-C, which falls closely on the curve for the unfertilized eggs.
In some of the experiments, for reasons yet obscure, over 80 per
cent of the unfertilized eggs remained immature after three hours in
the respirometers although controls in Syracuse watch glasses gave a
high percentage of maturation. Like the mature eggs, their rate of
oxygen consumption is constant, and for the first hour in the respirom-
eters it is equal to those of the mature and fertilized eggs, becoming
slightly lower after the second hour.
The absolute rates of oxygen consumption (Qo^) for these eggs of
the second series during the first hour in the respirometers expressed in
terms of cubic millimeters per hour per million eggs (the number being
obtained by hemocytometer counts) are: immature, 168; mature, 170;
and fertilized, 167. Thus it appears that the rate of oxygen consump-
tion for the Asterias eggs is the same whether mature, immature, or
fertilized. Their rate is of the same order of magnitude as that for the
fertilized Arbacia eggs, and is five times that of the unfertilized (Tang,
1931). If we take into consideration the diameter of the unfertilized
Arbacia eggs (74 micra) and that of the unfertilized Asterias eggs (160
micra), we obtain a ratio of 1 : 2.2. On squaring, it becomes 1 : 4.8,
which is the ratio of the Qo2, of these eggs, indicating that when ex-
pressed in terms of amount of oxygen consumed per unit surface, the
Qo2 of the two unfertilized eggs agree. Such a relation fails to hold
in the case of the fertilized eggs.
I wish to express my sincere gratitude to Professors R. S. Lillie and
R. W. Gerard for their advice and suggestions during the course of
this study.
CITATIONS
LOEB, J., AND H. WASTENEYS, 1912. Arch, entw.-mech. Organism., 35: 555.
TANG, P. S., 1931. Biol. Bull., 60: 242.
WARBURG, O., 1926. tiber Stoffwechsel des Tiimoren. Berlin. Julius Springer.
NOTES ON THE FEEDING MECHANISM AND ON INTES-
TINAL RESPIRATION IN CH^TOITKKUS
VARIOPEDATUS l
G. H. FAULKNER
( /•><>;» //.'(• Marine fiiohf/ical Laboratory, Woods Hole, ^fass.)
A healthy duct opt cms introduced into a glass tube rapidly lines this
with a parchment-like secretion. One individual, after living- in such a
tube for two weeks, extended the lining beyond the aperture of the tube
at one end. The prolongation was sharply constricted, showed suc-
cessive thickened rings, and terminated in an expanded rim; it was, in
fact, an exaggeration of the constriction at the end of a normal tube.
The tube current in such a preparation, as is well known, enters the
tube anteriorly and leaves posteriorly, maintained by the rhythmic beat-
ing of the fans on segments 14, 15, and 16. It is weak ventral to the
animal, but strong dorsally, and is directed under the arch formed by
tin- long parapodia of segment 12.
This main tube current provides the food supply, the nature of
which has been described by Enders (1909). The collecting mechanism
has been described by several authors as follows: the broad ciliated
buccal funnel collects directly from an extensive antero-ventral field;
in addition, ciliated grooves on the dorsal side of the thorax collect from
the tube current. To demonstrate this when the animal is removed
from its tube, food particles must be supplied by dropping them onto
the thorax from a pipette.
Such particle's are collected in a mucoid stream into grooves along
the inner edge of the arch formed by the parapodia of segment !_', and
pass from this anteriorly in a median groove. Particles which happen
to fall on the ventral face of the thorax are passed in laterally moving
^treams dorsally, between some of the posterior parapodia -being thus
lirought into the dorsal collect ing field. The median dorsal groove does
not lead directly into the month, but ends blindly in a dilatation posterior
to the dorsal lip overhanging the month i l-'ig. 1). The wall round the
terminal dilatation is thickened and raised, and forms a three-lobed
prominence. The anatomical details of the .structure of the groove
have been given by Jo\ en\-l .alTuie (1890).
1 Tin- followin \vi-rc made during a visit i<> tlu- Marine I'.inlogical Lab-
itory at \Vnnrls 1 1. .!(• .Inriii',' August and SepU-n '•,'•. 1"_N. The author1 wishes
tn thank Dr. !•'. K. l.illic \«r his interest and assistant-.
472
INTESTINAL RESPIRATION IN CH^TOPTERUS 473
In describing the transference of food from the groove into the
mouth, Enclers stated that " the lip of the buccal funnel is drawn back-
wards, and the ciliary groove, which now extends beyond the dorsal
border of the mouth, permits the granules to fall directly upon the ven-
tral lip of the funnel." Described in more detail, the complete course
of events is as follows.
While the food is passing forward in the groove, the anterior edge
of the dorsal lip is reflected posteriorly until its tip comes into contact
with the wall of the terminal dilatation of the groove (Fig. 2). To aid
this, the posterior half of the lip is depressed by ventral muscular con-
tractions centering in two areas. One of these is immediately anterior
to the end of the groove ; the other forms a pit within the tissue of the
lip. These two contractions result in the formation of a deep transverse
groove between the anterior end of the dorsal groove and the anterior
edge of the lip, arched over by the lip when this is reflected.
A further contraction now follows, as a result of which the exposed
surface of the reflected lip becomes depressed in the median-sagittal line
so as to form a deep longitudinal groove, which is a direct continuation
of the groove on the thorax ; the food particles can now pass from one
to the other without any interruption or obstruction (Fig. 3).
When the food has passed over the groove and into the mouth, the
lip is relaxed and returns to its position of rest. If it happens that the
food is removed before it reaches the anterior end of the thorax, the lip
does not complete this normal cycle of action, but is relaxed at once.
The stimulus which excites this reflex is apparently the presence of
solid particles in the food groove. In addition to this mechanical sensi-
tivity there may be some sense of chemical discrimination also, as the
animals often discard carmine or other non-nutritive particles.
The lip action can be induced experimentally in the following man-
ner. A fine brush from which all but a few hairs have been removed
is drawn slowly along the groove from the posterior end, and the lip
responds as described above: the advantage of using such a type of
stimulation is that one point only of the groove is stimulated at any one
moment. While the brush is in the posterior end of the groove there
is no response, but when it reaches approximately the level of the third
or fourth setigerous parapodium, reflection of the lip begins. The exact
extent of the anterior sensitive area varies, but it seems to be not more
than one-fifth of the total length of the groove. The lip is reflected
before the food reaches it, — it acts at such a time, in fact, that when the
first granules reach the end of the thoracic groove, the groove on the lip
is just ready to receive them. Stimulation of the anterior raised termi-
nation of the groove causes immediate response irrespective of whether
or not the groove itself has been stimulated previously.
474
G. H. FAULKNER
.
FIG. 1.
FIG. 2.
FIG. 3.
.. 1. Anterior end, dorsal view, showing lip in position of rest. (Diagram-
matic.)
2. The same, with lip reflected posteriorly.
! CG. 3. Lip rdk't U-<1 and grooved longitudinally ready to receive the food
MI 'Mm from tlie thoracic groove.
INTESTINAL RESPIRATION IN CH^TOPTERUS 475
In connection with this reflex action it is of interest to compare a
figure given by Joyeux-Laffuie of the nervous system of Chcetoptcrus.
He shows a pair of nerves arising from the dorsal region of the circum-
cesophageal ring and extending ovi-r approximately the anterior half of
the setigerous thoracic segments — thus corresponding more or less in
their distribution with the extent of the sensitive area.
Observations on intestinal respiration were made on individuals
which had recently regenerated some posterior segments. Such new
somites are transparent and free from pigment, and are particularly
favorable for this purpose..
Stephenson (1913) mentions Chcstopterus in his paper on intestinal
respiration and records an in-going current at the anus, but adds that
no anti-peristaltic contraction of the gut was seen. Such contractions
have, however, been seen repeatedly in recently regenerated somites,
though not in normal pigmented individuals. In addition to anti-
peristaltic contractions, the " gulping " action recorded by Stephenson
in several genera was seen at times, and there was in some cases also
observed a pulsating or pumping action in the gut at some distance in
front of the anus.
The simple anti-peristaltic action will be described first. It is an
anteriorly moving wave of contraction passing over the alimentary canal
in the few hind somites, constricting both the walls and the lumen. The
number of segments over which it persists varies, but it has repeatedly
been watched over at least seven segments, and occasionally over one or
two more. The interval of time separating successive waves varies also,
both in different individuals, and in the same individual on different oc-
casions : in fact, the activity often ceases altogether. When active, the
waves may follow each other at intervals of 4, 3, 2, or even 1% seconds.
A regular " gulping " action was seen only rarely, though it is not
uncommon to see the anus opening and closing at irregular intervals ;
this action is usually associated with a movement of protrusion and re-
traction of the posterior end of the canal. In one particularly favorable
individual the " gulping " action maintained a rhythm with intervals of
approximately one second, while after every three or four gulps there
was a pause while a peristaltic wave passed anteriorly over a few seg-
ments.
The pumping mechanism mentioned above probably serves to re-
inforce the peristaltic wave : it may synchronize with the wave, or may
have an independent rhythm. It is seen less frequently than the peri-
stalsis. The action occurs about seven somites in front of the hind end,
but as details vary, a few precise examples will be given.
One individual examined had seven newly regenerated somites at
the hind end, all perfectly colorless and transparent. Anti-peristaltic
476 G. H. FAULKNER
waves passed over the alimentary canal in the posterior segments, suc-
ceeding each other at intervals of approximately four seconds. At the
-aine time, the gut in the fifth segment from the hind end maintained a
pulsation independent of this wave, the beats occurring at intervals of
about one second. In another case peristaltic wave- parsed forwards
over the gut, and as they reached the seventh segment from the hind end
and were becoming weak, they received renewed impetus and persisted
through two or three segments further.
FIG. 4. Posterior end, showing the intestine protruded at anus.
The function of this anal and intestinal mechanism may be two- fold,
as suggested by previous authors (see Stephenson. 1913 and 1930).
In the first place, the in-going anal current may be respiratory: CIuc-
toptcnis has no special respirator}- organs, and then' are several features
which support the sngge-tion that the anus may play a part in respira-
tion. In the present case all the observations were made in aquaria,
hence, although the aeration was maintained as efficiently as possible, it
was not normal. IIowc\er. it i\ known that in the natural situation the
animal often protrn<le> its hind end from the tube. Further than this,
there is a terminal swelling on the alimentary canal which is protrusible,
and which, when exerted, forms a rosette-shaped protrusion around the
anus (Fig. I), \\heii retracted, the termination of the canal appears
compressed and much folded. There is also, as described by Enders. a
longitudinal groove in tin- inte-tiue in which the cells are distinguished
by their stronger cilia and by the absence of green granules. In the
oligocha-tes similar grooves are associated with an in-going respiratory
current, and the same explanation may perhaps be true here.
In the second place, the muscular activity of the intestinal wall may
h<- for the purpose of propulsion of blood in a peri-enteric sinus or
plexus: such a peri enteric plexus exists in Cluetoptcrus according to
Probst (1929).
INTESTINAL RESPIRATION IN CH/ETOPTERUS 477
SUMMARY
1. The food of Chatoptcrus is transferred from the dorsal thoracic
groove to the mouth by the temporary adaptation of the dorsal lip to
form a conducting channel leading directly into the mouth from the
blind anterior termination of the groove.
2. This reaction of the dorsal lip can be induced by mechanical
stimulation of the anterior part of the dorsal groo\r.
3. Clear and colorless somites which have been regenerated recently
at the hind end of a Chatopterus demonstrate the occurrence of anti-
peristaltic contractions in the alimentary canal of the hind segments;
such individuals also show a " gulping " action at the anus, and an ac-
cessory pumping mechanism in the walls of the intestine amplifying the
peristaltic contractions.
BIBLIOGRAPHY
ENDERS, H. E., 1909. A Study of the Life History and Habits of Chaetopterus
variopedatus, Renier et Claparede. Jour. Morph., 20: 479.
JOYEUX-LAFFUIE, J., 1890. fitude monographique du Chetoptere (Chsetopterus
variopedatus, Renier). Arch. Zool. E.rpcr., ser. 2, 8: 245.
PROBST, G., 1929. Das Blutgefiissystem von Chsetopterus variopedatus Renier.
Pub. Stat. Zool. Napoli, 9: 317.
STEPHENSON, J., 1913. On Intestinal Respiration in Annelids ; with Considerations
on the Origin and Evolution of the Vascular System in That Group.
Trans. Roy. Soc. Edin., 49: 735.
STEPHENSON, J., 1930. The Oligochzeta. Oxford University Press.
DIPLOID MALE PARTS IN GYNANDROMORPHS OF
HABROBRACON
P. W. WHITING
DEPARTMENT OF ZOOLOGY, UNIVERSITY OF PITTSBURGH
There are many theories of the origin of male parts of gynandro-
morphs in Hymenoptera. They may he classified under three headings,
n< genetic, androgenetic, and hiparental.
The gynogeiietic theories presuppose egg hinuclearity. Male parts
ari-e fnun a blastomere nucleus (Boveri, 1915), a separate oogonial
nucleus (Donhoff, I860; Doncaster, 1914), or a second ootid from the
tie oocyte (Whiting, P. W., 1924). According to these theories male
parts should show only maternal characters.
The theory of androgenesis involves polysprrmy i Morgan, 1905).
The supernumerary sperm nucleus undergoes cleavage resulting in
haploid male tissue. Male parts then should show only paternal char-
art'
The hiparental theory that holds for the majority of gynandromorphs
in J)r<>sopliilii ha- been applied hy Morgan to the bee. According to this
theory, the gynandromorph starts out as a female but loses an X-chromo-
soiue in an early emhrynnic stage. The resulting one X tissue is male.
Male tissue would then he hiparental in inheritance of autosumal traits,
hut as reijard- SCX-linked trails there is an equal chance that male tissue
would be of paternal or of maternal origin according to which X-chromo-
some was 1<
Another biparental theorv assumes both egg binuclearity and poly-
spermy. Fertili/atinn of the two egg nuclei by separate .sperm may
result in tissues <>f opposite sex depending upon the chromosome com-
po-ition df the two /.ygote-. In case of female digametism, male tissue
woulrl he entirely hiparental hut in case of male digametism, male tissue
\\-ould lie hiparental for autosomal traits, matroclim >us for sex-linked
characti
l're\ioiisly published records of gynaiidroiiu.rph- in ILibrobracon
have made it seem highly probable that male tissue is gynogenetic in
this form. A single case (No. 325) was, however, reported in which
male parts were patmclinous (Whiting, 1*. W., 1928). This example
had clearly male head and ocelli of male si/e. which as well as the eyes
were black and of paternal origin since the mother had recessive ivory.
478
DIPLOIDISM IN GYNANDROMORPHS OF HABROBRACON 479
A second instance came to light in August, 1930, when Mr. Hurst
Shoemaker was studying progeny from crosses of females from an
orange-eyed, o, defective-veined, d, stock (No. 3) with males of type
stock No. 1. Among the type females and orange-defective males ex-
pected from this cross, there was found a gynundromorph (No. 438)
with male head, black eyes and black ocelli. The ocelli were of typical
male size set in a dark area of male character. The antennae were also
male and the instincts were in general male ; for it attempted to mate
with females and was indifferent to host caterpillars, except for a slight
momentary reversal when it attempted to use its sting against a cater-
pillar. The abdomen was entirely female, body pigment and wings sym-
metrical. The primary wings were of normal venation, a patroclinous
trait but sex of wings in this instance, presumably female, could not be
accurately determined.
Further evidence has been obtained in regard to the nature of
gynandromorphs which bears upon the theories above presented. The
following summary involves only those with parents bearing diverse
traits,1 so that character of male structures is decisive as regards origin.
Of gynandromorphs from mothers carrying the dominant factor
there were four in which male parts were matroclinous. These are
decisive against the androgenetic theory for this case, but do not preclude
a biparental origin.
Of gynandromorphs from mothers carrying the recessive factor there
are 38 in which male parts were matroclinous. Among these the total
number of matroclinous traits in male parts is 50. These instances are
not only contrary to the androgenetic theory but against the biparental
theories as well.
The significance of the two individuals with male parts patroclinous
from recessive mothers will be discussed below.
Female parts of sex mosaics have generally been regarded as bi-
parental, and should accordingly show the dominant traits of either or
both parents. Four gynandromorphs obtained from mothers with a
dominant factor have shown this dominant. Twenty-eight obtained
from mothers with one or more recessive factors have shown 33 domi-
nant patroclinous traits, each dependent upon a single genie difference.
Evidence is entirely in agreement with biparental origin of female parts.
The reason for the excess of gynandromorphs from recessive mothers
and dominant fathers over those from the reciprocal is merely that many
more crosses are made in which the female bears the recessive. There
is no greater tendency for recessive females to produce them. Females
1 Many of the mutant factors causing these traits arose in the course of X-
radiation experiments conducted under a grant from the Committee on Effects of
Radiation on Living Organisms, National Research Council.
480 P. W. WHITING
beam, r m»rc rcce»i\es arc used in connection with investigations
on hiparcntal males.
(/n»scs of certain stocks regularly produce a few males resembling
their sisters in >howing the dominant traits of both parent-. (Whiting,
Anna R., 1927). These have been called at various times, anomalous,
patroclinous. l.i])arental, or diploid males. Evidence ha> heen gradually
accumulated which indicates their diploidism. It is perhaps useless to
speculate at this time as to why they are males if diploid ; but it has been
shown that occurrence of these males is dependent upon the stock of
mother as well as of father. Thus stock No. 3 female by related No. 1
male produces biparental males while the same female by unrelated Xo.
11 male fails to produce them. No. 11 males may. however, sire bi-
parental sons when crossed with related No. 12 females. Jt is suggested
that absence of an X-chromosome either from the reduced egg or from
the sperm may be the determining factor, but for this there is as yet no
evidence.
The two v,\nandromorphs with male parts patroclinous may be ex-
plained by the theory of loss of an X-chromosome in development but,
since both came from crosses producing biparental males, they are re-
garded as having developed from binucleate eggs in which each nucleus
wa> fertili/.cd by a different sperm. Egg binuclcarity and dispermy are
both involved with absence of an X-chromosome either from one e
nucleus or from one sperm nucleus.
LITI.RATURE CITED
BOVKKI, Tir.. 1915. I'li.-r die Kntstelmng cler Eugsterschen Zwitterbienen. Arch,
f. 7i;//,v. < h-iHinisin., 41: 264.
DOXCASTKK, I,., l(>\4. On the Relations between Chromosom limited Trans-
mission, and Sex-determination in .Ihni.rns grossulariata. J<ntr. Genetics,
4: 1.
I >• iXi.ii. Beitr. 3. Bii-iic>ikii>i<lc I (''her Z\\ en I'.ienen/.eitnnu.
.MoK'.AN. 'I'. }]., 1905. An Alternative Interpretation of the Origin »i Gynandro-
••p];oii, In-ccts. Science, 21: 632.
\Y i \.\ K., 1927. Genetic Hvidence for Diploid Males in Habrobraron.
ol. Bull.. 53: 438.
\Yiin i :.•.. i'. \Y., l''J4. Some Anomalies in 1 ladrobraron and their I'earin.^ Oil
.Maturation, l;ertili/.ation, and Cleavage. Am. Zool. Al>>tr. 140,
Anat. Rec., 29: 14-..
\VIIITI--.'-, 1'. \\'., l''_'s. Alo.saicism and Mutation in Habrobracon. J-linl. Bull.,
54: .
WHITING, 1'. \\'., AND ANNA R. WHITING, 1927. Gynandromorphs and Other Ir-
ular Types in f lahrohracon. Bio!. Bull., 52: 89.
A GYNANDROMORPH OF HABROBRACON FROM A
POST-REDUCED BINUCLEATE EGG1
P. W. WHITING AND MILTON FRANKLIN STANCATI
UNIVERSITY OF PITTSBURGH
The origin of gynandromorphs from binucleate eggs has been estab-
lished genetically for various insects. Boveri (Boveri, Th., 1915) re-
garded the two nuclei as resulting from a first cleavage division of a
reduced egg nucleus with consequent equality of maternal contribution
to male and female parts of the resulting embryo. His theory may be
called post-maturational.
Whiting (Whiting, P. W., 1924) interpreted the origin of a haploid
mosaic male from a heterozygous mother, oDzvM/OdWm, as due to
pre-reduction of Dd and Wzv, post-reduction of Oo and Mm. The two
cleavage nuclei would then be products of the second oocyte division,
one corresponding to the reduced egg nucleus, the other to the second
polar body. This maturational theory was later (Whiting, P. W., and
Whiting, "Anna R., 1927, and Whiting, P. W., 1928) applied to the
origin of gynandromorphs. Contrary to the view of Boveri, the two
ootid nuclei may bear different genes for those loci undergoing post-
reduction.
Other theories, maturational and pre-maturational, have been ad-
vanced by various authors allowing difference of maternal contribution.
Goldschmidt (Goldschmidt, R., 1931) has genetic evidence for the ex-
istence of such differences in the silkworm, Bombyx, as well as cyto-
logical i-esults favoring Whiting's maturation theory.
There is now abundant genetic evidence in Habrobacon for the ex-
istence of differences in the maternal contribution to the genetically
different parts of haploid mosaic males from heterozygous virgin
mothers. It has been supposed that gynandromorphs have an origin
similar to these males except that one of the ootid nuclei is fertilized
and that consequently female parts are diploid and biparental, while
male parts are haploid and matroclinous. There has been, however,
up to the present time no critical case in this wasp contrary to Boveri's
scheme.
1 The gynandromorph discussed in this paper was found during the course of
experiments conducted under a grant from the Committee on Effects of Radiation
on Living Organisms, National Research Council.
32 481
482
!'. \V. WHITIXG AND M. F. STANCATI
In the course of experiments at the Marine Biological Laboratory,
Woods llok-. during the summer of 1931, a number of females (stock
Xo. 3), homo/ygous for the recessive genes for orange eyes, o, and de-
fective wing venation, d. were crossed to type (black-eyed. 0, normal-
winged. /)) males (stock Xo. 1). \Yhcn the dihetero/.vgous, OoDd,
type (laughters from this cross were bred, the occurrence of females
among their progeny indicated that they had mated with their orange-
defective brothers. In addition to the four classes of males and of
females expected, — type, orange, defective and orange defective, — there
appeared in one fraternity a gynandromorph, Xo. 513 (Fig. 1).
FIG. 1. X2\.
The antenna.- were male; the left having _'l srgmcnts more or less
deficient terminally, the right having 23 oi normal appearance. Both
were orange in anterior and dorsal regions, black in posterior and
ventral. The ocelli (Fig. 2) were small and therefore female, the lat-
eral orange, tin- median containing some dark pigment. The area be-
tween the median and right ocelli was dark while that around the left
was yellow. The dark area may be presumed to lie male' in constitution,
as male integument tends to be darker than that of the female under
iiilar condition, of temperature, etc. The fact that the ocelli were
orange and female indicates that the orange parts of the compound
GYNANDROMORPH OF HABROBRACON 483
eyes as well were female and the black parts male. The left primary
wing was smaller, therefore presumably male, and showed defective
venation, the fourth branch of the radius (/?,) being completely lacking,
while the right primary was larger (female) and type. The secondary
wings also showed the sex difference in size. In the prosterna the left
FIG. 2. X 160.
side showed the darker (male) pigmentation. The abdomen was fe-
male throughout.
The origin of this gynandromorph may be represented by the fol-
lowing formula:
First polar body OoDd
Ootids Od
Sperm nucleus
oD
od
Cleavae
nuclei
The fact has been well established for Habrobracon that the mother con-
tributes to both male and female parts of gynandromorphs. The cir-
cumstances of this case, in which the paternal genes were recessive for
the loci concerned, allowing the dominant genes of the mother to express
themselves, indicate that the maternal contributions to the male and
female parts of this gynandromorph are different. That direct evidence
of this sort, in favor of the hypothesis of Goldschmidt and Whiting, has
not been found before in Habrobracon may be attributed to the fact that
most of the crosses are made with homozygous females, and that in the
few gynandromorphs reported from heterozygous mothers, the distribu-
tion of the haploid and diploid tissues did not permit differences between
the maternal contributions to show. By making enough appropriate
crosses with females heterozygous for factors affecting various parts of
the body, it should be possible to produce gynandromorphs giving fur-
ther evidence of the same sort.
LITERATURE CITED
BOVERI, TH., 1915. Uber die Entstehung der Eugsterschen Zwitterbienen. Arch,
f. Enhv. Organismcn, 41: 264.
484 P. W. WHITING AND M. F. STANCATI
GOLPSCIIMIDT, R., 1931. Die Sexuellen Zwischenstufen. Julius Springer, Berlin.
Pages 437^45.
\VHITINC. P. W., 1924. Some Anomalies in Habrobracon and their Bearing on
Maturation, Fertilization and Cleavage. Anat. Rec., 29: 146.
\Y: P. W., 1928. Mosaicism and Mutation in Habrobracon. Biol. Bull., 54:
289.
WHITIXG, P. W., AND ANNA R. WHITING, 1927. Gynandromorphs and other Ir-
regular Types in Habrobracon. Biol. Bull., 52: 89.
ON CERTAIN PHYSIOLOGICAL DIFFERENCES BETWEEN
DIFFERENT PREPARATIONS OF SO-CALLED
"CHEMICALLY PURE" SODIUM CHLORIDE
MARY MORRISON WILLIAMS AND M. II. JACOBS
(From the Marine Biological Laboratory, Woods Hole, Massachusetts, and the
Department of Physiology of the University of Pennsylvania)
I
It is the purpose of the present paper to direct the attention of
biologists to important differences in the toxicity to living cells and
organisms of certain commercial brands of so-called C.P. sodium
chloride which have usually been treated in the past as being more or
less identical chemically. The brands in question have all been used
frequently at the Marine Biological Laboratory and other scientific in-
stitutions in this country; and, in view of the striking differences that
will be shown to exist between them, the question arises how far the
work of different investigators, who have in the past used sodium
chloride of unspecified origin, is comparable and, indeed, how far many
published statements concerning the physiological properties of this
salt in pure solutions may be generally true. While these questions
cannot as yet be answered with entire certainty, the necessity is clearly
indicated for much greater care in the future than has been exercised
in the past in physiological work involving this commonest of all salts.
The observations which formed the beginning of this investigation
were made more or less accidentally in connection with certain unpub-
lished studies on the hemolytic effects of ammonium chloride on the
erythrocytes of the various classes of vertebrates, particularly the
fishes. In the course of these studies, controls of isotonic NaCl were
used for comparison, the salt employed being that which happened at
the time to be in general use at the Marine Biological Laboratory. It
soon became apparent that whereas the erythrocytes of the mammals
remained intact almost indefinitely in such control solutions, those of
several species of fishes, among them the sea robin, the butterfish, the
cunner, the tautog, the mackerel, the scup and the fresh water perch,
underwent destruction in times ranging from a few minutes to several
hours, though failing to do so in similar solutions of KC1 or CaCl2 or in
properly diluted sea water.
The unique behavior of NaCl is brought out in Fig. 1, in which are
485
486
M. M. WILLIAMS AND M. H. JACOBS
plotted against the times in hours from the beginning of the experiment
the cell counts, obtained by the usual hemocytometer method, of sus-
pensions of the erythrocytes of the sea robin (Prionotus carolinus) in
approximately isotonic solutions of KC1, NaCl and CaClo and in a
physiologically balanced mixture of the three salts. The rapid de-
struction of the erythrocytes here shown in solutions containing NaCl
and their preservation in the other solutions are entirely typical of
dozens of experiments made during the summer of 1926 with the par-
5,000-
0 L
12345
FlG. 1 . 1C fleet of exposing erythrocytes of the sea robin (Prionotus carol in its] to :
(1) M/3.7 KC1, (2) M/5.5 CaCl2, (3) M/3.7 XaCl and (4) a mixture of these solutions
in the proportions of 2:2: 96. Ordinates represent numbers of cells per cubic
millimeter and abscissa; times in hours.
ticular brand of salt in question, not only on the erythrocytes of the sea
robin but on those of the other species mentioned above as well.
On repeating the experiments the following year our surprise \vas
great when the expected hemolysis in Nad solutions completely failed
to appear, the erythrocytes remaining intact in such solutions for many
In mi-, \\iih no more evidence of injury than \\lu-n KC1 or properly
diluted sea water was employed. The only difference between the two
sets of experiments was that by chance a new brand of C. P. NaCl had
been substituted in 1927 for that used in 1(U6. On going back to the
l»i incr brand the earlier results could again be repeated at will. Kvi-
TOXICITY OF SODIUM CHLORIDE 487
dently there was in respect to their hemolytic properties at least, a very
decided difference between two preparations of NaCl, both presumably
of good quality and both in common use at the Marine Biological
Laboratory and elsewhere. Because of the possible importance of such
differences in physiological work, further experiments on fish erythro-
cytes were therefore undertaken with the more common commercial
brands of C.P. sodium chloride; and the results were later extended to
several other types of living material. The general outcome of these
experiments may now be described.
II
In all, five brands of C.P. NaCl, each prepared by a different manu-
facturer, were studied. In every case, samples from several separate
and previously unopened containers were used. In order to avoid any
possibly unjust conclusions being drawn as to the relative values of the
salts of the different manufacturers for the chemical purposes for which
they were primarily intended, the different brands will be designated
merely by the letters A to E, inclusive. It is perhaps not improper to
say that the brand designated by A, which is the least harmful to fish
erythrocytes of all those studied, being in fact practically as harmless
as KC1, is the Kahlbaum salt of the best quality obtainable. Of the
other four brands, B was at times almost as good as the Kahlbaum
preparation, but at other times was distinctly harmful, the differences
observed depending partly on the lot of salt used and especially on the
species of fish furnishing the erythrocytes. In our earlier experiments,
in which the decidedly resistant erythrocytes of the sea robin were
employed, this brand was almost indistinguishable from A, but in
later observations made by Dr. A. K. Parpart, working with one of the
authors on another problem, it appeared that it was quite incapable of
preserving for any length of time the much less resistant erythrocytes of
the tautog and the cunner which were, however, not markedly injured
by brand A. Brands D and E were invariably destructive to all the
fish erythrocytes studied, though more rapidly so to some than to
others. Brand C, as far as it was studied, appeared to be relatively
harmless, but our information about it is not very complete.
A typical experiment in which the effects on the erythrocytes of the
scup (Stenotomiis chrysops) of brands B, C, D and E and of KC1 is illus-
trated in Fig. 2. The blood in this case, as in all others here reported,
was freshly obtained from a living fish without the use of any anti-
coagulant and was added immediately to the solutions in question in
the proportion of approximately 1 to 200 by volume (i.e., one drop to
10 cc.). A slight variation in the sizes of the drops of blood was of no
• -
M. M. WILLIAMS AND M. H. JACOBS
-ince cell counts were made in every case. In the absence
of ' information concerning the osmotic pressures of the various
Hood.- -tudied, the concentration of NaCl employed was taken, unless
erwise indicated, as 0.25 M. Such solutions have a freezing point
»f approximately —0.86° C., which is not very far removed from that
of the plasma of the various marine teleost fishes for which figures are
,t\ailable; and at all events the concentration was the same for the
various brands of salt employed, so that the results were entirely
Comparable among themselves.
It will be noted in Fig. 2 that for the duration of the experiment
2,000 -
1,000-
0
FIG. 2. Effect of exposing erythrocytes of the scup (Stenotomus chrysops)to
M/4 solutions of brands B, C, D and E of NaCl and to M/4 KC1. Ordinates repre-
sent numbers of cells per cubic millimeter and abscissae times in hours.
(5 hours) there was no appreciable decrease in tin- number of erythro-
cytes in solutions of brands B and C, while in similar solutions of brands
n and /•.' the numbers had decreased very appreciably within one half
hour and very few erythrocytes remained at the end of four hours. It
may be mentioned incidentally that the erythrocytes of the scup, like
those of the sea robin, are relatively resistant ones; those of the butter-
fish or of the cunner disappear far more rapidly.
This particular experiment is typical of several dozen others differ-
ing in detail but all giving essentially similar results. In addition,
many incidental observations by \Y. A. Smith, S. K. Mill and A. K.
I'arpart working with one of the authors on other problems in which
cell counts were not made but hemolysis was followed by a macroscopic
TOXICITY OF SODIUM CHLORIDE
489
method have been in entire agreement with the results pictured in
Fig. 2. It may, therefore, be considered as definitely established that
the erythrocytes of certain fishes are affected in an entirely different
manner by various preparations of C.P. NaCl in common use.
Ill
As to the cause of these differences, two main possibilities suggest
themselves: either pure sodium chloride is in itself destructive to the
erythrocytes and its harmful effect is antagonized by impurities of some
sort present in brands A and usually in brands B and C, or pure sodium
chloride is in itself relatively harmless to this form of material and the
injury is due to a toxic impurity of some sort in brands D and E and
sometimes in B and C. Though the first type of explanation might
perhaps appear to be somewhat far-fetched, it must not be forgotten
that pure NaCl has been generally considered to be highly toxic (Loeb,
TABLE I
Effect on erythrocytes of the freshwater perch of solutions of NaCl of brands B and E
before and after recrystallization. The figures represent numbers of erythrocytes in 1
cubic millimeter of a dilute suspension.
Number of
Experiment
Brand B
Brand D
Original Salt
Recrystallized Salt
Original Salt
Recrystallized Salt
Beginning
of Experi-
ment
After
1
hour
Beginning
of Experi-
ment
After
1
hour
Beginning
of Experi-
ment
After
1
hour
Beginning
of Experi-
ment
After
1
hour
1
245
200
200
190
225
0
240
0
2
230
225
225
205
175
0
150
0
3
200
250
250
160
150
0
150
0
Average
225
225
225
218
183
0
180
0
1900; Osterhout, 1922) and that its toxic effects may be antagonized by
very low concentrations of plurivalent cations — for example, in the case
of the cilia of Mytilus, according to Lillie (1906), M/51,200 FeCl3 is
strikingly antitoxic.
It was thought that some light might be thrown upon these two
alternative types of explanation by a comparison of the effects upon
the same material of a harmful and a harmless brand of salt before and
after recrystallization. Brand B was known, for example, to be harm-
less and brand E to be highly destructive to the erythrocytes of the
freshwater perch. If the first of the two types of explanation were
correct, recrystallization should tend to make brand B more toxic than
before and leave E unchanged; if, on the other hand, the second were
correct, recrystallization should make E less toxic and leave B un-
changed.
490 M. M. WILLIAMS AND M. H. JACOBS
1 M k of time prevented extensive recrystallizations from being
carried out, but in Table I are represented the results of one experiment
in triplicate of this sort. Because of the difference in the osmotic
pressure of the blood of freshwater as compared \vith marine teleosts,
tin- concentration of NaCl here employed was 0.147 M, which has a
free/ing point in the vicinity of that found by Carrey il(>lo) for the
Hood of a number of freshwater fishes, i.e., approximate! \- -0.50° C.
It will be noted that the experiment shows no significant change in the
properties of either salt after recrystallization. It is therefore incon-
clusive, so far as throwing light upon the nature of the differences in the
physiological properties of the salt preparations in question is con-
cerned, but it does indicate one fact of great practical importance,
namely, that any impurities that may be present are difficult to remove
by recrystallization.
The question of possible antagonism was more directly and exten-
sively attacked in another way. Since it is well known from the work
of Loeb and others that perhaps the most effective single antagonist of
the toxic effects of sodium is calcium, and that solutions containing
sodium, calcium and potassium in the proper proportions form for most
cells and tissues a fairly good substitute for their natural medium, at-
tempts \\cre made to find some combination of the chlorides of calcium
or of calcium and potassium, with the toxic brands of sodium chloride
that would remove or at least greatly diminish the hemolytic effect of
the latter. In this we were completely unsuccessful. In particular,
the addition to the toxic brands of XaCl of Cadi and KC1 in the ap-
proximate proportions in which they occur in the body fluids of the
vertebrates or in sea water was almost without effect (see Fig. 1).
Only when isotonic solutions of CaCl2 or KC1 or both were added to
similar solutions of NaCl in sufficient quantities to dilute the latter
appreciably did a diminution of the hemolylic effect become apparent.
This effect, however, which is entirely different from antagonism, is
\\liat would be expected if the NaCl carried a toxic impurity.
It has been mentioned above that the erythrocytes of marine fishes
are preserved fairly normally in proper! v diluted sea \\ater, which is a
well-known example of a physiologically balanced salt mixture. In
several experiments, diluted sea water \\as mixed in different propor-
tions with approximately isotonic solutions ol one of the toxic brands
of XaCl. In such experiments it was found that the hemolytic effect of
t lie added salt could, in general, be detected to an extent that depended
ui)on its concentration in the mixture. This result is again what would
be expected if a toxic impurity were associated with the sodium
chloride.
TOXICITY OF SODIUM CHLORIDE 491
In view of the fact that all attempts to demonstrate a physiological
antagonism between the toxic brands of NaCl and various calcium and
potassium mixtures failed completely, the view was definitely aband-
oned that hemolysis by some salt preparations is due to the destructive
effects of pure NaCl itself. The fact that brand B could manifest its
harmful effects even in the presence of a considerable excess of diluted
sea water and the additional fact that brand A , which has been con-
sistently harmless, is at the same time one generally considered by
chemists to be of especially high purity seem to point rather to some-
thing added to the sodium chloride in the toxic samples of the salt.
It may be mentioned that Dr. Eric G. Ball has recently obtained evi-
dence of a very direct and convincing nature that the hemolytic effects
of some brands of NaCl are due to contained impurities. This evidence
will soon be published elsewhere.
Accepting the view that some brands of C.P. NaCl contain an im-
purity highly destructive to the erythrocytes of fishes there may be
mentioned briefly several of our unsuccessful attempts to determine the
nature of this impurity. Partly because our results on this point were
completely negative and partly because of the much more extensive
observations along the same lines soon to be published by Dr. Ball, it
will be sufficient here merely to eliminate from further consideration
one or two conceivable factors.
It is known that the erythrocyte is, in general, fairly sensitive to
pH changes and also that some preparations of so-called "neutral
salts" are not entirely neutral. One of the first of the possibilities to
be considered, therefore, was the reaction of the various solutions stud-
ied. It was found that as far as pH measurements can be made upon
completely unbuffered solutions there were no significant differences in
reaction between the different sodium chloride solutions and the dis-
tilled water used to make them up, or between these solutions and simi-
lar ones of completely harmless KC1. Furthermore, in one experiment
there was added to solutions of brands B and E sodium bicarbonate in
the proportion of one part of M/4 bicarbonate to twenty of M/4 NaCl.
The pH of the resulting mixtures was then adjusted to 7.0 in each case
by the addition of carbon dioxide in the proper amounts, a procedure
which leaves the effective osmotic pressure of the mixture for the
erythrocyte unaltered. Blood was then added to these well buffered
mixtures, which were kept tightly stoppered throughout the remainder
of the experiment. In spite of this careful regulation of the pH of the
solutions, the erythrocytes underwent destruction in the presence of
NaCl of brand E and remained intact in the case of brand B exactly as
before. In still other experiments, it was shown that with a given
492 M. M. WILLIAMS AND M. H. JACOBS
• ,1 rli.in^c in the reaction of the solutions of two pH units (i.e., from
[ill (>.o i" N.OK which greatly exceeds any differences that could con-
•. ably have been present in any of our experiments, had negligible
ects upon the characteristic properties of the salts. It may be con-
sidered fairly certain, therefore, that the physiological differences be-
tween the salts in question are not due to pH effects.
In our search for possible impurities in sodium chloride preparations
it was suggested to us by a chemical colleague that fluorides, which are
fairly toxic to some living cells, might perhaps be concerned. Experi-
ments were therefore made in which sodium fluoride was added in
different proportions to tin- harmless salt of brand B. The proportions
used ranged from a maximum concentration of NaF of 0.025 M by a
series of dilutions with a factor of one fifth to a minimum concentration
of the order of 0.00000001 A I. In none of these solutions, however, was
brand B caused to resemble even remotely brands D and E, and it was
therefore concluded that fluorides could scarcely be the impurity con-
cerned. Similar experiments were carried out with salts of several
toxic metals such as Pb, Hg and C'u which might conceivably have been
present in traces in the more injurious salt preparations, but our results
were again essentially negative.
As far as it was possible to carry our experiments up to the time
when it became necessary to discontinue them in 1928, absolutely no
clue had been obtained as to the nature of the hypothetical impurity.
It should be emphasized, however, that a lack of knowledge of the
nature of this impurity in no wise detracts from its physiological im-
portance or renders its disturbing effects in certain types of experi-
mental work less real.
IV
After establishing the fact that certain brands of so-called ('.I'.
NaCl are highly destructive to the erythrocytes of a number of teleost
fishes, experiments were undertaken to determine how far similar
effects could be obtained with other forms of living material. It is evi-
dent that effects of this sort might, if unrecognized, cause considerable
confusion in physiological work, particularly since all the brands of
sodium chloride in question have been commonly used in such work-
frequently with no published statements by which they may be identi-
1. Additional experiments were therefore undertaken upon the
following forms of material: mammalian erythrocytes, newly-hatched
l-'undulns, the eggs of Arbacia and the cilia. of Mytilus. These experi-
ments may be briefly described in the order mentioned.
As contrasted with the erythrocytes of the fishes, those of the mam-
inals appear to be little injured by any of the brands of sodium chloride
TOXICITY OF SODIUM CHLORIDE 493
in question. It is doubtless owing to the comparative insensitiveness
of this much-studied type of cell that the striking physiological differ-
ences in the properties of different salt preparations did not long ago
become generally known. Our experiments were carried out on the
blood of man, the ox, the dog, the cat, and the porpoise in the manner
described above, the only difference in technique being that the con-
centration of the salt employed with the mammalian erythrocytes was
0.1 54M instead of 0.25 M.
The results in the case of every mammal studied were, briefly, that
for 10 or more hours at room temperature or for 24 hours partly at
room temperature and partly in a refrigerator there was no appreciable
hemolysis in any of the solutions. In experiments of longer duration,
there were in a few cases some slight indications of differences in the
expected direction, but these were so small and irregular as to be of
little significance. It is possible that by employing aseptic precau-
tions, which were not practicable in our experiments, and by keeping
the erythrocyte suspensions for several days, constant differences might
be demonstrated. For practical purposes, however, in ordinary ex-
periments of short duration with mammalian blood it would appear to
make little difference which brand of sodium chloride is used.
The experiments made upon Fundulus heteroclitus are of interest
because it was upon this material that Loeb (1900 and later papers)
obtained his most striking evidence of the toxicity of pure sodium chlo-
ride. Though for a number of reasons it appeared to be impossible
that the effects described by Loeb could have been due to an impurity
in the salt used rather than to the salt itself, it nevertheless seemed of
some importance to determine whether with Fundulus the primary
toxicity of pure sodium chloride might be modified in any way by the
contaminating impurity supposed to be present in some preparations.
The general result of our experiments was to show that this is, in fact,
the case.
A typical experiment on newly-hatched, free-swimming fish is de-
scribed in Fig. 3. In it, brands B and E were compared with respect
to their ability to stop (a) the swimming movements of the animals
and (b) the heart-beat. The concentration was in each case M/2,
which is approximately isosmotic with Woods Hole sea water. It will
be observed that the differences between the two salts are rather
striking. At the end of 6 hours nearly all swimming movements had
ceased in the animals exposed to brand E, while only a few of the indi-
viduals exposed to brand B had been similarly affected ; some continued
to move in this solution for over 12 hours. The cessation of the heart-
beat also occurred much more rapidly in the presence of brand E than
494
M. M. WILLIAMS AND M. H. JACOBS
in that of brand B.> These differences were observed many times with
no exceptions. It may be concluded, therefore, that the observed ef-
fects of sodium chloride upon Fundulus depend to a considerable extent
en the particular salt preparation employed.
A \ cry sensitive test object for many purposes is the egg of Arbacia,
whose rate of cleavage is affected in a readily measurable manner by
very slight changes in, for example, the osmotic pressure and the carbon
dioxide tension of the surrounding medium. Since pure isotonic
sodium chloride is known to be toxic to this egg, it was thought that
differences in the properties of different salt preparations might be
100
\
B
20
4 8 12 16 20 24
FIG. 3. Effect on newly-hatched Fnndulus of M/2 XaCl of Brands B and E.
Ordinates represent percentages of the animals showing swimming movements
(solid lines) and heart-beat (broken lines); absciss;e represent times in hours.
shown by exposing fertilized eggs of Arbaci,i to (hem for suitable times
and then determining the effect of such exposures upon the rate or the
tinal percentage of cleavage. This was done in two ways : first by plac-
ing the ev,gs in (he sodium chloride solution to be tested shortly before
i leavage and allouing them to remain in (he solution, and second by
employing a short temporary exposure to the salt followed by a return
to sea \\ater. The first type of experiment proved to be entirely un-
suitable owing to the failure of the eggs to di\ ide at all, but the second
t\pe yielded results which, while not wholly satisfactory, were at least
suggestive.
TOXICITY OF SODIUM CHLORIDE
495
The general result obtained from experiments of this latter type was
that in some cases there were no very significant differences between
the effects of salts B and E, but that in several cases where decided
differences appeared, these were always in such a direction as to indi-
cate a greater toxicity of brand E than of brand B. The reverse condi-
tion was never obtained. An experiment showing a very considerable
difference in the toxicity of salts of brands B and E is summarized in
Table II. It may scarcely be considered a typical experiment, how-
ever, since the differences observed were usually not so great.
Finally, a few observations were made upon the cilia of Mytilus,
which Lillie (1906) has shown to be very rapidly injured in solutions of
pure isotonic sodium chloride. It is, of course, difficult to treat the
beat of cilia in a strictly quantitative manner, since different groups
TABLE II
Effect on Subsequent Cleavage of Exposure of Fertilized Arbacia Eggs to Two Brands of
Sodium Chloride
Percentage Undergoing Cleavage within 2 Hours
Length of Exposure
Brand B
Brand E
minutes
5
68
3
10
39
4
15
4
1
20
2
5
25
8
2
come to rest at different times and even within the same group certain
individual cilia continue to beat long after the others have ceased to
do so. It is possible, therefore, for the experimenter merely to esti-
mate in a general way when some given end-point has been reached.
As far as such estimates could be made, our experiments showed no
significant differences between the different brands of sodium chloride,
perhaps because in this case the pure salt itself is extremely toxic. For
these particular experiments brand A was not available, but a sample
of B of very low toxicity to fish erythrocytes was compared with brand
E of high toxicity. The times for the attainment of the same estimated
end-point with different gill filaments were found to be 15, 13, 4.5, 11
and 9.5 minutes (average 10.6 minutes) with brand B; and 11, 13, 7 and
10.5 minutes (average 10.4 minutes) with brand E, respectively. It is
not impossible that more extensive and refined experiments would be
capable of demonstrating definite differences, but as far as the present
evidence goes, these would not likely be very great.
496 M. M. WILLIAMS AND M. H. JACOBS
summari/ing the results of the various experiments with different
types of living material, it may be said that no significant differences be-
tween the different brands of sodium chloride studied have been found
with mammalian erythrocytes or with the cilia of Mytilus; occasional
Inn by no means constant differences in the expected direction have
been found with the eggs of Arbacia, constant differences of consider-
able magnitude in the same direction occur with newly-hatched Fundu-
Ins; and differences of the most striking and characteristic sort are in-
variably present in the case of the material which was first studied,
namely, the erythrocytes of certain fishes. Though up to the present
time the fish erythrocyte is the most sensitive form of material known,
it is not impossible that other types will be discovered in the future of
even greater sensitivity. In the meantime, physiologists should con-
stantly be on their guard, when working with sodium chloride, against
what, at its worst, is capable of being a source of serious experimental
errors.
SUMMARY
1 . Of five commercial brands of C.P. sodium chloride that have
been studied, one is apparently always harmless and two always de-
structive to the erythrocytes of certain teleost fishes; one and perhaps
both of the others are intermediate and somewhat more variable in
their properties.
2. There is indirect evidence that the destructive effect of the toxic
brands is due to the presence of an impurity of some sort, which has,
however, not been identified. It is not removed by a single recrystal-
lization of the salt.
3. Similar though much less striking differences have been found in
the physiological action of the brands of sodium chloride in question
upon newly-hatched Fundulus and less certainly upon the eggs of
Arbacia. No constant differences have been noted in the case of
mammalian erythrocytes or in that of the cilia of Mytilns.
4. It is suggested that in all physiological work in which sodium
chloride is used particular attention should be given to the possibility
of errors resulting from the presence in the --.ill of unknown toxic
impurities.
BIBLIOGRAPHY
GAKKI-.V, \V. I-:., 1916. Am. Jour. Physiol., 39: 31.v
1. 1 i.i. 1 1., K. S., 1906. Am. Jour. Physiol., 17: 89.
LOEB, .[., 1900. Am. Jour. Physiol., 3: 327.
Os'ii i HMI i, \V. J. \ ., \'>22. Injury, Recovery and Death, in Relation to Conduc-
tivity and I Vrnirability. Philadelphia.
SPECIFIC INFLUENCE OF THE HOST ON THE LIGHT
RESPONSES OF PARASITIC \VATKR MITES
JOHN H. WELSH
THE ZOOLOGICAL LABORATORY, HARVARD UNIVERSITY
The common fresh-water mite Uninnicola ypsilophorus var. haldc-
n/aiii (Piers), which lives as a parasite between the gills of the mussel
Anodonta cataracta Say, exhibits interesting modifications in its behavior
to light associated with its parasitic life. Mites which are removed
from the influence of material from the host show a positive response
to light, but when a small amount of water from the mantle cavity or an
extract of gills of the host is added to water containing the mites there
is an immediate and striking reversal to a negative state. A negative
response to light is necessary to keep the parasites within the host and
it was suggested (Welsh, 1930) that this reversal in light response
might be considered adaptive and secondarily acquired. An attempt
was made to determine the nature of the substance which causes the
reversal and it was concluded that certain proteins or decomposition
products of proteins were responsible for the reversal, which perhaps
is something of the nature of a conditioned response with olfactory or
taste organs being involved in the conditioning. In order to test this
idea, in part, it was necessary to determine whether or not material
from the host was specific in causing the reversal in the light responses
of the mite. The present paper is concerned with the results of this
investigation.
The majority of tests were made on the mites from Anodonta.
although two other species from other fresh-water mussels were also
tested. The experiment was simple and consisted merely in comparing
the effect of water from the mantle cavity or a water extract of the
gills of several different fresh-water mussels on the behavior to light
of mites from a given host. Tests on the mites from Anodonta were
made with material from the following fresh-water bivalves.
Lainpsilis radiata (Gmelin) 1 Houghton's Pond
Elliptic complanatus (Dillwyn) j Blue Hill, Mass.
Sphaerium siilcatuui Belmont, Mass.
497
33
498 JOHN H. WELSH
Cycltnuiis tnbcrcnlata (Raf.)
r.nrynla iris (Lea) Huron River,
Liyninia fasciola (Raf.) Delhi, Mich.
/:///7>//o dilatains (Raf.)
The tests were made under constant conditions of temperature and
intensity of illumination, as it was found that both of these factors
influenced the behavior of the mites. In a typical experiment six mites
\\ere placed in each of two small rectangular glass dishes 8 cm. long,
J cm. wide, and 2 cm. high. The mites had previously been washed
in several changes of water in order to remove any host material and
had been free from the influence of the host for at least a day. The
small jars containing the mites were placed in a large water bath where
the temperature was maintained at approximately 18° C. The source
of illumination was a 6 volt, 18 ampere ribbon filament lamp at a
distance of thirty centimeters from the end of the tank. A glass
window permitted the light to enter the tank, where it passed through
twenty centimeters of water before it readied the jars containing the
mites. The mites were always kept in 5 cc. of water and the gill
extract prepared by grinding the gills in distilled water and then filter-
ing, or the water from the mantle cavity, was added in 1 cc. quantities
at the end of the jar towards the light. \Yhene\er a reversal occurred
the mites moved out of the region of extract faster than the diffusion
took place.
Following is the record of a typical experiment:
April 22. Placed six mite-, in each of two jars ./ and />'. All positive
to light.
3:40. Added 1 cc. of extract of gills of (.'I'dainiis to . /. One mite
stimulated in some way, and temporarily indifferent to light.
returned in 30 seconds to light end of jar.
3:41. Added 1 cc. of extract of gills of Anuilnutd to jar />'. Five
mites immediately negative, one mite unaffected temporarily by
extract.
3:45. All mites in A positive. All mites in />' negative.
4: 00. No chan
4:30. Removed and washed both lots of mites and returned to same
jars in fresh water.
1:35. Added 1 cc. of extract of Anodoiil.i gills to jar A. All mites
immediately and actively negative.
4:40. Added 1 cc. of extract of Cyclonais gills to jar B. No indica-
tion of a response, all mites remaining positive.
This experiment was typical of all the tests made on the mites from
Anodonta with material from each of the seven other species of mus-
LIGHT RESPONSES OF WATER MITE 499
sels used. Gill extract or water from the mantle cavity of Anodonta
only, was effective in bringing about a reversal in the light response <>l
Unionicola from Anodonta.
Two other species of Unionicola were found, Unionicola fossitlata
(Koen) from Cyclonals tubcrculata, and an undetermined species from
Lawpsilis radlata. Both species of mites were tested with material
from their hosts as compared with extract from Anodonta. They both
showed negative reactions to the same light intensity used in the tests
on the mites from Anodonta but were found to be positive to a low
light intensity obtained by using a neutral tint filter transmitting 0.1
per cent of the light used in the previous tests. A reversal in their
light response could be brought about by material from their own host
but not by material from Anodonta.
These results indicate that the material present in the host which
causes a reversal in the light responses of parasitic water mites is
specific for a particular host-parasite combination. It is possible that
certain water mites have more than one host, as do certain parasitic
copepods, in which case one would not expect to find the same specific
influence of host on parasite. However, the majority of parasites are
always found associated with a particular species as host and the
results of these tests help to explain why this is true. A long con-
tinued parasitic or commensal life tends to modify an animal struc-
turally and at the same time to modify the behavior of the animal and,
as was pointed out in an earlier paper (Welsh, 1930), these mites
which are found in Anodonta were probably primitively positive to
light and the negative response has been acquired only after a long
period of life within a particular species of mussel. The constant
influence of some material from the host which stimulates the mites
either through their olfactory or gustatory organs keeps them in a
negative state as regards their behavior to light of a given intensity,
but their removal from this influence causes a reversion to the primitive
state.
The evidence thus far indicates a specific response on the part of
fresh-water mites to some material from their host, and it is expected
that similar associations exist in other host-parasite combinations. A
study of these associations should reveal interesting modifications in
behavior and yield further information regarding the intimate physio-
logical relationships existing between closely associated animals.
LITERATURE CITED
WELSH, J. H., 1930. Reversal of Phototropism in a Parasitic Water Mite. Biol.
Bull, 59: 165.
IS OSMOTIC HEMOLYSIS AN ALL-OR-NONE
PHENOMENON ?
ARTHUR K. PARPART
(Prom the Department of Physiology, University of Pennsylvania)
Tin-re are at present two opposed views concerning the relation be-
tween the disappearance of the erythrocyte and the amount of hemo-
globin which it liberates during osmotic hemolysis. Some investigators
(Ruzynyak. S.. 1911; von Lieberman and von Fenyvessy, 1912; and
J. Baron, 1928) believe that the disappearance is a gradual process
resulting from a slow escape of hemoglobin, and hence that swollen
though visible cells may have lost considerable quantities of this sub-
-tance. Baron, for example, has reported that hemolysis induced by
hypotonic salt solution may lead, as in oiu- experiment which he cites,
to the disappearance of 17 per cent of the cells while at the same time
the amount of hemoglobin recovered in the surrounding solution is as
high as 42 per cent. The conclusion is drawn that even visible cells
must have lost a part of their hemoglobin. Baron and others have
designated hemolytic processes of this type a> " partial.'
Another group of investigators (Dienes. I... 1911; II. Handovsky,
1912; S. C. Brooks, 1918; and G. Saslow, 1928-29) hold that the
erythrocvte disappears with great rapidity at tin- time of hemolysis.
Brooks, who worked chiefly with licniol\si> by ultra-violet radiation,
has summari/cd this concept in the following way: " When hemoglobin
finally begins to diffuse from a given ervthrocyte. the process is so
quickly completed that it may ordinarily be regarded as instantaneous."
According to this view the process is of the type customarily termed
" all-or-none." A parallel situation is believed by many workers to be
found in osmotic hemolysis. It is held by such workers that the cell
subjected to osmotic changes which result in swelling, does not begin
to lose- its hemoglobin until it has attained a fairly definite volume,
termed the " hemolytic volume," at which time a rapid outward diffu-
sion of hemoglobin occurs to a point of equilibrium with the surround-
ing fluid.
The apparent " hemolytic \oluiuc" is believed to differ for different
crythrocvtes, even in blood from the same individual, hence the well-
500
OSMOTIC HEMOLYSIS 501
known range of resistance always encountered in experiments on os-
motic hemolysis. In general, according to the second view that hemoly-
sis is an all-or-none process, it ought to he possihle to find a solution of
such a concentration as to cause, in a given sample of blood, a complete
loss of hemoglobin from all the cells whose resistance falls below a
certain value, without any loss from the cells of higher resistance. Ac-
cording to the first view, however, that partial hemolysis is a phenome-
non of common occurrence, such a sharp separation should not be
possible.
The difference between these two views is of more than theoretical
interest. If the first one is correct then little significance can be at-
tached to the term " percentage hemolysis," and standards such as those
used, for example, by Jacobs (1930), while reproducible and therefore
of practical value, correspond to nothing encountered in actual experi-
ments. If, on the other hand, hemolysis is actually an all-or-none phe-
nomenon, then such standards represent not merely the apparent but
also the true percentage of hemolysis, with a consequent gain in the
significance of the results obtained.
The study of the relation between the disappearance of the cell and
the liberation of hemoglobin involves measurements of two sorts: first,
a count of the number of cells in the sample of blood employed, and
second, a determination of the total hemoglobin content of the cells fol-
lowed by an estimation after hemolysis has occurred of the number of
cells undestroyed and of the hemoglobin content either of these cells
or of the supernatant fluid, or preferably of both. An obvious point,
which has been neglected by previous workers in securing ideal condi-
tions for the outward diffusion of hemoglobin, is the use of a large
volume of surrounding solution relative to the total volume of the cells,
so that a true diffusion equilibrium would permit the escape of almost
all of the hemoglobin that is free to leave the affected cells.
Under conditions where the outward passage of hemoglobin is not
limited by an insufficient external volume, a comparison of the number
of cells destroyed and the amount of hemoglobin liberated might con-
ceivably lead to any one of three following results: (1) The amount
of hemoglobin in the surrounding solution corresponds exactly to that
contained in the cells that have disappeared. The process is therefore
" all-or-none " in character. (2) The amount of hemoglobin in the
surrounding solution is greater than that contained in the cells that have
disappeared. This would suggest a " partial " process, at least in the
case of some of the cells. (3) The amount of hemoglobin in the sur-
rounding fluid is less than that contained in the cells that have disap-
peared. Such a condition would indicate a slight retention of hemo-
502
ARTHUR K. PARPART
liin by imisible cells. All of these possibilities are illustrated in
I by data selected from the previous literature or obtained in the
course i if this investigation.
Strictly speaking, the " retention" type must always be present to
• \tcnt when hemolysis occurs in finite volumes of solution since a
diffusion process can do no more than bring about an equality of distri-
bution of hemoglobin between the cells and their surroundings. In
previous work, where the whole blood was introduced into the hemolytic
solution in the volume ratios of 1 to 1 to 1 to 5 (Baron, J., 1928; G.
Saslow. I1 '28-29) this retention must of necessity have been great. In
the present work, however, this effect has been minimized by employing
a ratio of 1 to 2.000, thus pro\iding an opportunity for an almost com-
plete outward diffusion of the hemoglobin.
TABLE I
Summary of the Possible Types of Osmotic Hemolysis
Type
Us
I >; sap-
red
1 li-inc I'J'ibin
App<
llcino]\-<i> in
rvei
Maximum
1 i lor of
\lrt hod
(1)
"all-or-none"
f>er
50
/><r cent
51
I )iliiu- plasma
Saslow
per cent
7
75
76
\.i( 1, ^1\ ri'fol,
clll\ -li-nc -lyrnl
author
4
(2)
"partial"
17
12
1 )ilut<- plasma
Baron
}
(3)
retention
66
58
dl\i erol
ily Stag<
aut Imr
4
It should be noted further that a process of type (3) might be ex-
pected to occur as a temporary stage in the attainment of end results of
type ( 1 ). Whether or not it would be observed would depend on the
rapidity of tin- process and upon the' conditions governing the visibility
of the cells. Theoretically also a combination of types (2) and (3)
might conceivably at times simulate type ( 1 ). That such a combination
of effects would occur so exactly and consistently in an extended series
of experiments as to lead to erroneous conclusions is, however, ex-
tremely unlikely.
In the present investigation it has been found that the condition de-
ibed as type ( 1 ) is always present when the hemolytic system has
attained a final equilibrium. In certain cases, for example, during the
course of heniolysis produced by glycerol solutions, a temporary reten-
tion of some hemoglobin is exhibited during the early stages of the
OSMOTIC HEMOLYSIS 503
process ; but at final equilibrium the relation is the same as before,
namely, " all-or-none." No evidence whatever has been found under
the conditions of these experiments of the escape of hemoglobin from
visible cells; that is, of " partial hemolysis " of individual erythrocytes.
II
To determine accurately the relation between the number of erythro-
cytes that disappear and the amount of hemoglobin liberated in a given
hemolytic system it is necessary that a great number of careful cell
counts be made. In the present work these counts were made with the
usual counting chamber, the areas customarily used for white cells be-
ing employed, and at least 10 of these areas of 1.0 sq. mm. each being
counted. This procedure involved a count of from 1,500 to 2,000 cells
for each solution. By counting a large number of unit areas (160) and
also a large number of cells the accuracy of the cell count was increased
so that the maximum error was 2 per cent as determined by the method
of Student (1907). The use of dilute cell suspensions (1 to 2,000)
eliminated the difficulties encountered by previous workers owing to the
presence of so-called " ghost " cells. This point will be considered in
detail later, and has been touched upon here solely to stress the fact
that at no time during the course of these experiments did the difficulty
which previous investigators (Baron; Saslow) found in distinguishing
between " stromata " and intact cells arise.
The method of hemoglobin estimation employed in all of these ex-
periments was a new one particularly well adapted to this purpose. It
was worked out in conjunction with Dr. W. R. Amberson and Dr. D. R.
Stewart, and will be described in detail elsewhere (1931). The princi-
ple involved in the method is that of the optical pyrometer, and its most
valuable features are the high sensitivity and accuracy that can be ob-
tained. The sensitivity proved to be entirely adequate to deal with the
minute amounts of hemoglobin that the use of very dilute cell suspen-
sions necessitated. Thus, in experiments in which whole blood was
mixed with distilled water in the proportion of 1 to 2,000 it was found
that between this concentration and zero concentration of hemoglobin a
series of at least one hundred readings could be obtained by the pyrome-
ter. This represents a maximum error in the hemoglobin readings even
at these unusually low concentrations of 1 per cent. The term " per-
centage of hemoglobin " is everywhere used in this paper to designate
the amount of hemoglobin in the surrounding fluid of a given hemolytic
solution relative to that contained in the total number of cells employed.
It is known that whole blood on standing even for a short period of
time (2 to 3 hours) and at a low temperature (3° to 5° C.) often under-
504
ARTHUR K. PARPART
a slight degree of hemolysis which may range between 0 and 4 per
cuit. The amount of hemoglobin in the plasma introduced with the
whole blood into the hemolytic solutions, as well as other pigments that
may be present, must, therefore, be corrected for in the hemoglobin
determinations. This has been accomplished by diluting whole blood
in the proportions of 1 to 2.000 in isotonic saline and then rapidly (10
minutes) centrifuging the cells out. A pyrometer reading on the
supernatant fluid gives the error due to the hemoglobin and other pig-
ments in the plasma. This correction for plasma error has been applied
in all the experiments here recorded.
The dilution steps involved in the hemoglobin determination, the
preparation of the cell suspensions and the cell counts were found to
have an error of not more than 1 per cent. Since the cell counts in-
volved an error of approximately 2 per cent and the hemoglobin deter-
minations of not more than 1 per cent, the total maximum error in com-
parison of the disappearance of cells and of the appearance of hemo-
globin in the solution was of the order of magnitude of 4 per cent. It
will be noted that in Table II and in Figs. 1 and 2 the agreement is well
within this figure.
TABLE II
/><it<i lllnstrnttie of the Accitrncy of the Method
\ Percentage of hemoglobin appeared
3
3
12
20
25
30
51
62
77
95
B. 1'crccnta^c of hemoglobin in cells re
maining
99
96
89
78
78
70
50
;<)
21
4
\ • r,
102
• )<)
101
98
103
too
101
101
98
99
Table II also gives another very important check on the accuracy of
the method in demonstrating that the hemoglobin which was recoverable
! nun the visible cells plus the hemoglobin in the supernatant fluid is
e<|iii\ alnit. within the experimental error, to that contained in the
original number of cells. There is no evidence that this vcrv necessary
test has been made by previous workers and it is apparently to its omis-
sion that much of the confusion in the literature is attributable.
A point of particular importance in obtaining the results here re-
corded is strict attention to the influence of the factors discussed by
Jacobs and 1'arpart ( I'^l), of which temperature, pi I and the attain-
ment of a final equilibrium are the most important. A recent worker
( Saslow, ( r., 1'L'S 2() i states that " Most of the experiments performed
were failures because of the difficulties above enumerated: lack of con-
trol of degree of hemolysi-, and unsatisfactoriness of the cell count."
OSMOTIC HEMOLYSIS 505
Only in three instances was he able to secure a suitable degree of hemol-
ysis, and this unfortunately fell within the- narrow range of from 40 to
60 per cent. In the present work, by a careful control of the factors
influencing the degree of hemolysis, namely temperature, pH and the
attainment of equilibrium, and by the use of dilute cell suspensions, thus
allowing a practically complete outward diffusion of hemoglobin from
the affected cells which removes the necessity for determining " ghosts,"
it has been found possible to obtain very readily any desired degree of
hemolysis. This may be observed in Fig. 1, where the entire range
from zero to 100 per cent has been covered with gaps of no more than
5 per cent.
Ill
Osmotic hemolysis has been used throughout this work. The ex-
periments performed fall into two classes. Those belonging to the first
class involved the use of a non-penetrating substance, sodium chloride,
in hypotonic solutions of concentrations so chosen as to bring about a
disappearance of some but not all of the erythrocytes present in the
suspension. By varying the concentration in small steps it was pos-
sible to cover the entire range from zero to 100 per cent hemolysis.
After the hemolytic system had reached its final equilibrium condition
or after hemolysis had been checked in the manner described below, a
comparison was made between the number of cells that had disappeared
and the amount of hemoglobin that had been liberated. In the second
group of experiments hemolysis was allowed to occur in solutions, orig-
inally isosmotic with blood, of the penetrating substance glycerol and, in
a few cases, ethylene glycol. The hemolysis produced by the pene-
trating substances was checked at various points by the addition of
sodium chloride in the proper amount and comparisons were made as
before between the liberation of hemoglobin and the disappearance of
cells. Because of their greater simplicity, the experiments involving
solely the entrance of water into the cells will be described first, and
those involving the penetration of the solute as well will be dealt with
in the following section.
As has already been mentioned, the factors of temperature and pH
must be so regulated that their effect on the degree of hemolysis attained
is a constant one. The temperature of the hemolytic solutions employed
in these experiments was maintained at 20° ±0.1° C. by means of a
water bath. The pH of the hypotonic solutions was controlled by the
addition of a small amount of phosphate buffer. All solutions were
prepared from a stock solution consisting of 14 parts of molar NaCl
and 1 part of molar Na2HPO4, brought to a pH of 7.0 by the addition
506
ARTHUR K. PARPART
of a t: .oentrated HC1. Upon dilution of this stock solution to
concentrations between 0.5M and 0.05M, the pH of the resulting solu-
tions was 7. 4<> within the limits of accuracy of the quinhydrone electrode
(± 0.02). Since the pH of the hlood used is originally not far from
this point, the comparatively slight buffering of tin- solution is sufficient
for all ] Tactical purposes.
In the experiments involving the simplest type of osmotic hcmolysis
tin- procedure was as follows: To 50 cc. amounts of various hypotonic
salt solutions, usually differing from one another by 0.001 M, 25 cu. mm.
of whole blood, defibrinated by whipping, was added from a calibrated
hemoglobin pipette afu-r the solutions had been brought to a temperature
of 20° C. ± 0.1L, in a water bath. Following the addition of blood they
were' gently and continuously stirred, at the temperature stated, for a
TABLE III
Data on the Blood of One Animal (Ox) Illustrating the Applicability of the
All-or-none Concept at Equilibrium
n of
[PO<
A. Cells
I ) sappeared
1'.. 1 li •mo'ilnliin
Appeared *
B-A
Mil
l>rr cent
/••r cent
0.154
0
0
—
0.115
3
3
0
0.110
9
12
+3
0.100
22
23
+ 1
0.095
27
30
+3
0.090
64
62
_2
O.OS5
79
77
-1
0.080
92
95
+3
* After correction for the plasma error.
period of one hour, which is more than sufficient for the attainment of
equilibrium (Jacobs and Parpart, 1931). A small portion of the solu-
tion was then removed for the cell count, while the rest was centrifuged
at 2,000 r.p.m. for 15 minutes and the supernatant fluid siphoned off for
the hemoglobin estimation. The time allowed for centrifuging was
-liown to be sufficient by two procedures, namely, bv microscopic exam-
ination of the supernatant fluid which revealed the presence of no cells,
and by measurement with the optical pyrometer which gave the same
transmission values for the fluid whether it had been centrifuged for 10
minutes or for one hour.
Table 111 presents the results obtained in a tvpical experiment of
this sort with ox blood. The concentrations indicated represent dilu-
tions of the original stock solution containing both XaCl and NaL,HPO4
concentration was somewhat arbitrarily taken as unity. Though
OSMOTIC HEMOLYSIS
507
osmotically the concentrations given are not exactly equivalent to sim-
ilar ones of pure NaCl, the differences are very small and the solutions
are entirely reproducible. The values given in column A represent the
difference between the total number of cells employed and the number
of cells remaining when the hemolytic system was at equilibrium, and
are expressed on a percentage basis. Column B gives the percentage of
hemoglobin in the surrounding solution, which was determined after
correction for plasma error as previously described.
Inspection of Table III shows that the hemoglobin which appears in
the supernatant solution corresponds, within the limits of experimental
error, almost exactly with that originally contained in the cells that
have disappeared. There is no evidence, therefore, at the end of the
time in question (1 hour) of any appreciable retention of hemoglobin
nor of partial hemolysis of cells still visible. The process under these
conditions appears to be strictly " all-or-none " in character.
100
80
4)
<L>
a.
5 60
0)
U
(+H
O
a> 40
bo
a
V
CL,
20
0
0
100
20 40 60 80
Percentage of Hemoglobin Appeared
FIG. 1. The liberation of hemoglobin from ox erythrocytes in hypotonic NaCl
solutions under equilibrium conditions.
More extensive corroborative data are shown in Fig. 1, where the
results of 50 experiments in which the blood of 12 animals was used are
plotted. The solid diagonal line represents the condition that should
obtain if the hemolytic process, at equilibrium, is all-or-none in char-
50?
ARTHUR K. PARPART
acter. The experimental points scatter about this line within the error
of the method: that is, with a maximum deviation of 4 per cent and an
average deviation of 1.7 per cent. Since it is inconceivable that in such
a large number of experiments as exact an agreement as this could be
obtained by a fortuitous combination of retention of hemoglobin and
partial hemolysis, the conclusion seems inescapable that an all-or-
none process i> being dealt with.
The results and conclusions summari/ed in Tables II and III and
in Fig. 1 were- obtained with ox blood, while the conflicting data of
Haron ( l('2Si and Saslow (1928—29) were based ujxjn experiments on
human blood. Since there i> a marked difference in the osmotic re-
sistance of human blood and of ox blood it seemed advisable to perform
a number of similar experiments on the former type of blood. These
100
80
&
a
I 60
~
U
cd
•M
=
at
HI
CL,
40
20
0 JO 10 60 80 100
I 'en-enlace of J lemoglobin Appeared
I-'K.. J. Tlie liU-r;ition of hemoglobin from luiman crytlirocytes in liypotonic
NaCl under equilibrium conditions.
experiments are represented in Fig. 2, in which the diagonal line has
the same significance as before. It \\ill be noted from this figure that
the evidence of an all or-uone relationship is as unmistakable in the
case of human as in that of ox blood.
In all of the experiments so far described bemolysis has been al-
lowed to go to complete equilibrium. The question arises whether there
OSMOTIC HEMOLYSIS 509
is a similar agreement between the hemoglobin that has escaped and the
number of cells that have disappeared at times before the final equilib-
rium of the system has been attained. Stated in a more concrete man-
ner, if in a given hemolytic system the concentration is such that the
final equilibrium will involve the destruction of. for example, 75 per
cent of the cells present, will the all-or-none concept hold when only
25 per cent of the cells have undergone hemolysis?
To answer this question, the rate of hemolysis of ox blood in hypo-
tonic salt solutions in which at equilibrium there remained some per-
centage of the cells unhemolyzed, was first followed by the method of
Jacobs (1930), and in this way it was determined at what time after
setting up a given experiment any desired degree of apparent hemolysis
had been attained. ]t was then possible to repeat the experiment and
to stop the hemolysis at the chosen point and to make comparisons as
before. In a number of experiments of this type 25 cu. mm. of whole
blood of the ox was added to 25 cc. amounts of hypotonic saline solu-
tions whose hemolysis-time curves had been determined as above. At
the end of a definite time (15 seconds to 2 minutes) 25 cc. of salt
solution of a concentration to make the whole isotonic (0.154M) was
suddenly mixed with the hemolyzing solution, thus stopping hemolysis.
As before, the temperature was kept at 20° C. and the pH at 7.4. Cell
counts were then made on a portion of the solution in which hemolysis
had been stopped, while the remainder was centrifuged and the super-
natant fluid used for the hemoglobin estimations in the usual manner.
A series of these determinations led to the results recorded in Table IV.
It is evident from this table that the agreement between the per-
centage of the cells that have disappeared and the hemoglobin that has
been liberated is as good as it was in the cases where the final equilibrium
had been reached. The all-or-none concept, therefore, is not limited
merely to the end-stage of osmotic hemolysis of this type but probably
applies throughout the entire process. The conclusion would seem to
be warranted that in studies on the kinetics of osmotic hemolysis the
cell may be assumed to liberate all of its hemoglobin at the time of its
disappearance.
IV
Osmotic hemolysis in solutions of penetrating substances is a some-
what more complicated process than that so far described since the rate
of entrance into the cell of the solute as well as that of water is involved.
Since previous workers have apparently not studied the nature of the
hemolytic processes induced by these substances, it appeared of interest
to determine whether they might also be associated with an all-or-none
type of hemolysis. The substance chiefly studied, namely glycerol.
510
ARTHUR K. PARPART
was eho- "iily because it is relatively non-toxic, but because its
rate "f penetration into the erythrocyte of tbe ox is sufficiently slow so
that the hemolytic process can bf stopped at any desired point by the
addition of salt in proper concentration. A typical experiment may now
cribed.
TABLE IV
Effect of Stopping Hemolysis Before the Attainment of Equilibrium
n of
NaCl • NaiHPl >>
A. Cells
Disapi ie ired
B. lU'inoiJnbm
A]>]"
B-A
Mil
per / i-nl
per cent
0.0'ni
71
72
+ 1
0.0"!)
31
27
-4
0.090
36
35
-1
0.090
37
38
+ 1
0.088
75
74
-1
0.088
35
32
-3
0.088
45
43
-2
0.088
49
52
+3
O.I
67
64
-3
0.0
20
22
+2
0.085
33
29
-4
0.08S
40
40
0
0.0
so
78
_2
0.0
55
53
_2
0.0
57
58
+ 1
0.082
61
58
-3
Whole defibrinated Mood of an ox was introduced into an isosmotic
solution of glycerol, in the proportion of 25 cu. mm. of blood to 25 cc.
of 0.3M glycerol. The solution was gently stirred and kept at a tem-
perature of 20° C. and the rate at which hemolysis proceeded determined
by the method of J , • . r*30). I ntil about 35 minutes after setting
up such a system, no hemolysis was found I" occur; between 35 and
55 minutes the pn>ccs> proceeded fairly rapidly from 0 to 100 per cent,
it was then a simple matter in subsequent experiments to stop the
hemolysis at any desired point between these two time intervals by the
niion to 25 cc. of the suspension undergoing hemolysis of 25 cc. of
a solution at pi I 7.4 containing 0.308M NfaCl XaJll'O, and 0.3M
cerol. After the addition of this solution the whole was equilibrated
JO C. with stirring for a period of one hour, at which time cell counts
and estimations of the hemoglobin liberated were made in the manner
described in the previous section.
OSMOTIC HEMOLYSIS
511
Blood samples from five animals were tested in this manner and the
results are summarized in Fig. 3. In this figure, as in the two previous
ones, the diagonal line represents the theoretical result that should ob-
tain if the process is all-or-none. It will be observed that in this case
the experimental points in these determinations scatter on one side of
100
80
-o
fi
60
0)
U
i
&
20
0
0
80
100
20 40 60
Percentage of Hemoglobin Appeared
FIG. 3. The liberation of hemoglobin from ox erythrocytes during the early
stages of hemolysis by glycerol.
this line in the direction of an hemolytic process exhibiting a retention
of hemoglobin. The deviations, which amount to 5 to 12 per cent, are
too great to be accounted for by experimental errors alone.
Two possible explanations of these results suggest themselves. The
first is that the penetration of the glycerol might so alter the refractive
index of a number of the more swollen cells that they do not appear in
the count though they still retain all or a part of their hemoglobin. The
second is that the hemolytic system may not have attained its final
equilibrium at the time the observations were made.
As a test of the former possibility the following experiment was
performed. Whole blood of the ox was pipetted into a hypotonic saline
solution that would induce a slight degree of hemolysis, in the propor-
tion of 25 cu. mm. of blood in 50 cc. of salt solution at pH 7.4. These
solutions were equilibrated for one hour with gentle stirring, at 20° C.
512
ARTHUR K. PARPART
At the end of tin- equilibration period 25 cc. was removed for cell count
and hemoglobin estimation. To the remaining portion was added an
equal volume of a solution containing the same concentration of salt
plus O.i'M ijvccrol. Thus the salt concentration remained unchanged,
while the solution became isosmotic with respect to glycerol. This lat-
ter -olution was equilibrated for one hour in the same way, at which
time cell counts and hemoglobin determinations were a-ain performed.
I )ue allowance was made for the one-half dilution necessitated bv the
procedure. The results of several such experiments are presented in
Table V.
TABLI Y
f Glycerol mi the Refractive Index nf the Erythrocyte
Hi-l'.m- tin- A l.litiini di r.Kvi-rol
Alter tin- AiMitiun <if (ilyo-rcil
A. Cells
Disappeared
1'.. 1 Iciiioi;]ul)in
Appeared
B-A
A. Cells
Disappeared
I',. 1 Irinnvlnliin
Appeared
B A
per it-lit
per cent
per n'lil
pi-r i fill
35
32
-3
40
38
_2
31
29
-2
32
33
+ 1
28
29
+ 1
33
32
-1
50
51
+ 1
52
51
-1
Had the glycerol in any way affected the refractive index of the
corpuscles, then the data obtained after tin- addition of glycerol should
have departed from the all-or-nonc relationship by an amount com-
parable to that observed in Kig. 3. As no such shift was observed, it
may be concluded that -Ivcerol does not appreciably alter the refractive
index of the erythrocytes. The explanation, therefore', of the discrep-
ancy between the hemoglobin liberated and the cells disappeared in a
hemolytic system of the glycerol type would appear to lie aloni; the lines
of the second sux^estion. namely the time required lor the attainment
of equilibrium.
To test this point, whole blood of the ox was equilibrated lor one
and for four hours, respectively, in hypotonic saline solutions containing
varving concentrations of ^Ivcerol between 0.03M and 0.3M. As be-
fore, _}5 en. mm. of blood was added to 50 cc. of each solution. To
make the- experimental conditions comparable, half of the solution was
n-nio\ed for the determinations at the end of one hour, the rest being
equilibrated in the customary way for an additional period of three
hours. The concentrations of salt and of glycerol used are shown in
Table VI.
i .\aniination of this table shows that in the absence of glycerol, as
OSMOTIC HEMOLYSIS
513
in previous experiments, the hemoglobin liberated at the end of either
one or four hours agrees within the experimental error with that in the
cells which have disappeared. In the presence of glycerol, however, at
the end of one hour there is an apparent retention of hemoglobin similar
to that indicated in Fig. 3, whereas this had completely disappeared by
the end of four hours. In brief, it would appear that at the final equi-
librium the conditions are the same as before, but that the final equilib-
rium is longer in being attained.
TABLE VI
Production by Glycerol of an All-or-none Type of Hemolysis, at Equilibrium
Solution
After 1 Hour
After 4 Hours
A. Per-
centage
of Cells
Dis-
appeared
B. Per-
centage
of Hemo-
globin
Appeared
B-A
A. Per-
centage
of Cells
Dis-
appeared
B. Per-
centage
of Hemo-
globin
Appeared
B-A
0.090 M NaCl + Na2HPO4
90
88
2
89
88
-1
90
90
0
88
89
+ 1
Same + 0.03 M glycerol
68
60
-8
88
87
-1
Same + 0.10 M glycerol
66
58
-8
93
94
+ 1
Same + 0.30 M glycerol
34
28
-6
91
92
+ 1
As to the cause of this delay in the escape of hemoglobin, several
possibilities might be suggested. One of the most plausible is that the
retention of hemoglobin during the first part of the hemolytic process
may be due to a blocking of the hypothetical " pores " by which it is
frequently assumed to leave the swollen cells, either because of the rela-
tively large size of the glycerol molecule or because of its adsorption on
the walls of the pores. In either case, the effect of the glycerol would
be to decrease the surface area through which hemoglobin may leave
the cell. The data in Table VI lend confirmation to this view, for not
only does the presence of a very small amount of glycerol (0.03M)
markedly slow the rate of disappearance of the cells, but it also causes
marked temporary retention of hemoglobin. This is true, in spite of
the fact that the solution of 0.03M glycerol and of 0.09M NaCl is
osmotically equivalent to a 0.105M solution of NaCl, a concentration
which of itself rapidly produced about 15 per cent hemolysis. It may
further be determined from this table that the initial distance of the
hemolytic system from osmotic equilibrium does not influence the
34
514
ARTHUR K. PARPART
amount of retention. As has already been pointed out, the extent of
this retention is outside the limits of experimental error of the methods
here employed. It averages ahout 8 per cent as compared with a maxi-
mum experimental error of 4 per cent and, as will be observed in Fig. 3,
the deviation is consistently in the same direction.
As contrasted with glycerol, ethylene glycol, a closely related sub-
stance of lower molecular weight, fails to produce any observable reten-
tion of hemoglobin even during the early stages of the hemolytic process.
Its behavior is indicated by the data given in Table VII, which were
TABLE VII
Hemolysis by Ethylene Glycol
Concentration of
NaCl • \.i-IIPO4
in 0.3 M
Ethylene Glycol
A. Cells
Disappeared
B. Hemoglobin
Appeared
B-A
per cent
per cent
0.085
15
19
+4
0.085
19
20
+ 1
0.080
0.080
45
46
47
46
+ 2
0
0.078
60
58
-2
0.078
60
62
+2
0.075
0.075
75
76
76
76
+ 1
0
secured by equilibrating for one hour the usual dilution of ox blood in
0.3M ethylene glycol made up in the salt solutions of the concentration
indicated in the table. It will be noted that the presence of the ethylene
glycol affects neither the rapid attainment of equilibrium nor the all-or-
none character of the hemolysis that the hypotonic solutions alone would
have exhibited. Except for the IO\V<T molecular weight and molecular
volume there is no obvious reason why this substance should differ so
markedly from glycerol.
V
The results that have been obtained fail entirely to confirm Baron's
(1929) contention that osmotic hemolysis is a "partial'1 process.
I lemolysis of this type has constantly been found to be an all-or-none
phenomenon both under equilibrium conditions and while in progress,
except for the temporary retention of hemoglobin that occurs during
the early st;i-« s produced by a penetrating substance like glycerol.
OSMOTIC HEMOLYSIS 515
Even in this case, however, the end-point finally attained by the system
is the same as in that found under the simpler conditions where the
entrance of water alone is involved.
It was mentioned previously that Baron's results were secured by
the use of 1 : 1 to 1 :6 dilutions of whole blood with water. To make
cell counts and hemoglobin estimations he added sufficient hypertonic
salt solution to make the dilute hernolytic solution isotonic. After ob-
taining cell counts he then centrifuged the solution and made the hemo-
globin estimation on the supernatant fluid. The futility of such a
procedure might, however, have been recognized from the results of
Bayliss (1924-25), or of Adair, Barcroft and Bock (1921) and others.
These workers have clearly demonstrated that when whole blood is
diluted with water in proportions such as those mentioned above, the
corpuscles swell and lose hemoglobin only to a degree that corresponds
with a diffusion equilibrium between the cells and the external solution.
If at this point salt is added to make the solution isotonic, the cells
shrink, thus trapping in them sufficient hemoglobin so that they again
become visible. Bayliss (1924—25) discussed these conditions as fol-
lows: " Suppose the ghosts had a volume two and one-half times the
volume of the original corpuscles and contained the same concentration
of hemoglobin as the surrounding fluid. Then, if on shrinking to their
normal size they become more or less impermeable to hemoglobin, they
might contain finally a hemoglobin concentration some two to three
times that of the external fluid."
It would appear, therefore, that Baron, when making his cell counts,
must have included a number of cells whose visibility had been restored
after they had previously lost a portion of their hemoglobin. That the
hemoglobin was not more completely lost was obviously due to the small
volume of the external solution. Under these experimental conditions
it is not surprising that no exact correspondence could be obtained be-
tween the number of cells that had disappeared and the amount of hemo-
globin in the surrounding medium. As has been mentioned, Saslow
(1928-29), in his studies on hypotonic saline hemolysis, using the
method of Baron, obtained data which contradicted those of the latter
author. The reason for this discrepancy is perhaps furnished by Sas-
low's statement that before making the cell count he pipetted off and
discarded the " stromata." The so-called " stromata " were probably
cells which had lost their hemoglobin to equilibrium with the surround-
ing fluid and by discarding them he made his cell count on cells which
had not lost hemoglobin, and naturally the cell count corresponded with
the hemoglobin in those cells. If he had determined not only the
amount of hemoglobin in the cells undestroyed but also that in the sur-
5 If. ARTHUR K. PARPART
rounding fluid it is likely that the two amounts would have totalled con-
siderably less than 100 per cent. However, this crucial test was not
made.
An attempt has been made, by the use of Sallow's dilution of blood
with distilled water, to determine the hemoglobin content of the super-
natant fluid after " reversion " of the cells by means of sufficient NaCl
to make the solution isotonic followed by centri fusing at 5,000 r.p.m.
for one and three hours. A clear supernatant fluid is obtained in this
way. but further addition of salt to it causes it to become clouded, and
microscopic examination shows this cloudiness to be due to the re-
appearance of cells. If this latter cloudy supernatant fluid is allowed
to stand for some fifteen to thirty minutes it again becomes relatively
clear and can be made cloudy again by making it still more hypertonic
with salt. A short time later it again becomes clear. Thus, in the
method of Saslow, and also in that of Baron, the " clear " supernatant
fluid really must contain cells which have approximately the same spe-
cific gravity and, more important, the same hemoglobin content as the
surrounding fluid. Because of this circumstance hemoglobin determi-
nations on the supernatant liquid of very concentrated suspensions,
whether made directly, as by Baron (1928), or indirectly, as by Saslow
(1928-29), are unreliable.
Consideration of these facts leads to the conclusion that only when
conditions arc created such that the volume, of the cells at the time of
hemolysis is relatively small as compared with that of the surrounding
fluid, can the nature of hemolysis be determined. Dilution of whole
blood in hypotonic salt solution in the proportion of 1 : 2,000 adopted
in these experiments leads to practically infinite dilution of the hemo-
globin content of the hemolyzed cells. Under these conditions " ghost "
cells and the phenomenon of "reversion" introduce no complications
and it becomes possible to demonstrate that osmotic hemolysis is an
all-or-none phenomenon, and that the term " percentage hemolysis " has
a real significance.
SUMMARY
1. Hemolysis produced by hypotonic sodium chloride is of an all-
or-none tvpe, that is. hemoglobin either fails to escape from the erythro-
cyte or does so completely up to the point permitted by the attainment
of a diffusion equilibrium in the system.
2. The same all-or-none character is observed when, instead of
permitting tin- liemolytic process to proceed to its original equilibrium
position, it is stopped at an intermediate point by the addition of sodium
chloride.
OSMOTIC HEMOLYSIS 517
3. In hcmolysis in solutions of penetrating substances such as
glycerol and ethylene glycol, the final equilibrium obtained after check-
ing the process by the addition of sodium chloride likewise indicates an
all-or-none relationship.
4. In glycerol solutions the liberation of hemoglobin lags somewhat
behind the disappearance of the cells and the final equilibrium is rather
slowly attained.
I am greatly indebted to Dr. M. H. Jacobs for the suggestion of this
problem and for his helpful criticism.
BIBLIOGRAPHY
ADAIR, G. S., J. BARCROFT, AND A. V. BOCK, 1921. Jour. Phys'wl., 55: 332.
BARON, J., 1928. Pfliiger's Arch., 220: 242.
BAYLISS, L. E., 1924. Jour. Physiol, 59: 48.
BROOKS, S. C., 1918. Jour. Gen. Phvsiol., 1: 61-80.
DIENES, L., 1911. BiocJiem. Zcitschr., 33: 268-274.
HANDOVSKY, H., 1912. Arch, ex per. Path. u. Phann., 69: 412.
JACOBS, M. H., 1930. Biol Bull., 58: 104.
JACOBS, M. H., AND A. K. PARPART, 1931. Biol. Bull.. 60: 95.
PARPART, A. K., W. R. AMBERSON, AND D. R. STEWART, 1931. Biol. Bull.. 61: 518.
RUSZNYAK, S., 1911. Biochem. Zeitschr., 36: 394.
SASLOW, G., 1928-29. Quart. Jour. Exper. Physlol., 19: 329.
STUDENT, 1907. Biometrika, 5: 351.
VON LIEBERMANN, L., AND B. VON PENYVESsv, 1912. Zcitschr. I inuiunltiitsforscli.
Orig., 12: 417.
TIIK DETERMINATION OF HEMOGLOBIN' CONCENTRA-
TION IN DILUTE SOLUTIONS
ARTHUR K. PARPART, WILLIAM R. AMBERSON AND
DOROTHY R. STEWART
(From the Department of Physiology, University of Pennsylvania)
Colorimetric methods for hemoglobin determinations have been
found to have an error which varies between 1 and 5 per cent ( Schwent-
ker, F. F., 1929). The chemical procedures for the quantitative estima-
tion of hemoglobin, such as the measurement of the iron content (Fow-
weather, F. S., 1926) and the carbon monoxide capacity (Van Slyke,
D. D., and A. Hiller, 1928), have a greater accuracy, i.e., .5 to 1 per
cent. These methods, however, necessitate the use of hemoglobin solu-
tions of high concentration.
The investigator desiring to make a large number of hemoglobin
determinations in a reasonable time, and under circumstances where
only minute amounts of hemoglobin are available, must have recourse
to some other method. An optical system suggests itself, but it should
combine simplicity of operation with a high degree of accuracy.
The hemoglobin de-terminations reported in the preceding paper
(Parpart, 1931) were secured by a method which employs the principle
of the optical pyrometer. The equipment necessary is readily pro-
curable; the determinations may be made with great rapidity, and the
intrinsic error of the method is 1 per cent. A striking feature of this
optical system is its sensitivity for low hemoglobin concentrations. To
determine the sensitivity we have studied a solution of hemoglobin,
standardized by the iron content method, kindly furnished by Dr. \V. C.
Stadie, and have found that the apparatus used is capable of detecting
changes in hemoglobin concentration of the order of 1.2 X 10~5 mM.
(M. W.= =68,000) in the range from 1.3 mg. to 0 mg. per 50 cc. of
solution. In most uses of the method absolute values are not deter-
mined, since accurate relative values are sufficient.
The apparatus should also prove useful in quantitative determina-
tions on the related pigments in both animals and plants, especially
under circumstances where these pigments are procurable only in small
amount-,.
518
HEMOGLOBIN CONCENTRATION IN DILUTE SOLUTIONS 519
II
Our application of the pyrometer principle to this problem employs
an optical system in which the light intensity of a line filament in a
pyrometer lamp is matched against a diffuse background of fixed in-
tensity. Hemoglobin solutions of different concentrations are placed
in a 20 cm. polariscope tube between the two light sources. The ap-
paratus is similar to that described by Amberson (1922, I).
A lamp, termed the pyrometer lamp (C), is placed at a distance of
about 1 foot from another lamp, the background lamp (£). A small
telescope (B) fixed at its focal distance from the pyrometer lamp,
makes it possible to view a portion of the filament of this lamp against
a background of light emitted by the background lamp and rendered
diffuse by a ground glass plate set in at one end of the polariscope tube.
Details of this arrangement may be observed in Fig. 1.
B
/^\
^
tf 41 irv
1
mt
\
F
>*
tan
sr*\
x—A s^r^ 1 V — ?„
2.&A. 6 5 A.
FIG. 1. Diagram of the optical pyrometer as applied to the estimation of
hemoglobin concentration in dilute solutions. The symbols are explained in the
text.
The pyrometer lamp is a small 3-volt bulb with a U-shaped filament
of tungsten of about 0.2 mm. in diameter; while the background lamp
is a Mazda bulb rated at 26 volts, 6.6 amperes, 2,500 L. Both lamps
are run by storage batteries since line circuits fluctuate too much to
permit adequate control. Frequent observations are made of an am-
meter (G) in series with the background lamp circuit and its intensity
is maintained constant by adjustment of a coarse-fine parallel resistance
(a and b), in series. The current passing through the pyrometer lamp
is recorded by a milliammeter (G'), and can be varied in steps of about
0.25 milliampere by a coarse-fine parallel resistance (a' and b') in
series.
520 A. K. PAR I 'ART. \V. R. AMBERSON, AND D. R. STEWART
In theM- hemoglobin determinations it is not necessary to know the
relation existing between the intensity of the pyrometer lamp and the
current passing through it, since all readings are evaluated by reference
t<> a calibrated curve made with hemoglobin solutions of known con-
centration.
An essential feature in securing high sensitivity is the use of a green
glass filter (A) which is placed at the eyepiece of the telescope. The
transmission characteristics of this filter presented in Fig. 2 were ob-
tained spectrophotometrically through the kindness of Dr. D. L. Drab-
kin. It will be noticed that the filter has its maximum transmission in
0.10
600
500
450
0 550 a
Wave Length
I1" in. 2. Transmission curve for the green filter.
the region of maximum absorption bv hemoglobin, whether the latter
is in the oxygenated or reduced condition, or in the form of methemo-
globin or acid bematin.
A 20 cm. polariscope tube ( /' ). containing the hemoglobin solution,
is placed on a rigid stand ( /•" ) between the two lamps. The usual plate-
glass disc covers the end of the tube facing the pyrometer lamp, while
at the other end there is a glass disc ground on one surface. This latter
disc serves to diffuse the light from the background filament. Both of
these discs must always be replaced in their original position as rotation
will vary their transmission.
The readings ha\e been taken with the background lamp set and
stabili/ed at 4.7 amperes. \Vith the hemoglobin solution in place, the
i •*
amperage- of the pyrometer filament is varied until the portion selected
for observation just disappears, proceeding in every case from dark to
light. The milliamperagc of the pyrometer lamp is recorded and the
HEMOGLOBIN CONCENTRATION IN DILUTE SOLUTIONS 521
readings made in triplicate. After a brief amount of practice these
readings check within 0.5 to 1.0 milliamperes.
Ill
Due allowance must be made for certain variations in the optical
system and for other possible sources of error, and these will now be
considered. Slight changes in the brightness of the lamps and varia-
tions in transmission of the interposed glass surfaces have been observed
from week to week. These changes have been determined by taking
readings on distilled water before and after each experiment. Such
readings taken as much as 24 hours apart have always checked within
the limits of error of reading and hence the lamps and the transmission
of the glass may be considered constant for this period.
400
390
380
37o'
"i- r
/
/•"
— '
X
H
1
I
\ \
0
.01
.02
.03
.04 .10 .25
FIG. 3. Effect of salt concentration on the transmission of a dilute hemo-
globin solution (dilution 1 to 4,000). Readings were taken 20 hours after prep-
aration of the solution. Ordinates represent readings of milliammeter ; abscissae,
concentration of NaCl in mols per liter.
When very dilute solutions of hemoglobin are employed, a few de-
tails must be carefully controlled. The dilute hemoglobin solutions are
obtained by hemolyzing whole blood in distilled water in the proportion
1 to 2,000. At this dilution it can be shown that the absorption of light
is influenced by some factor or factors in addition to hemoglobin con-
centration itself, since the addition of salt increases the transmission.
Figure 3 shows the relation between salt concentration and transmission
(measured in milliamperes). The reason for the decreased transmis-
sion between 0.015M salt and distilled water is uncertain; that the pre-
cipitation of serum globulins may play a part in it seems possible, but
as their concentration is very small at such dilutions they probably do
not account for the entire effect. This factor has been taken into ac-
count in all determinations by making the hemoglobin solutions up to
522 A. K. PAR TART. W. R. AMBERSOX, AND D. R. STEWART
0.10M XaCl -f- N~a,IIPO4 (in the ratio 14 parts XaCl to 1 part
Xa,l I !'(.), i at pi I 7.40. All dilutions have been made with a salt
solution of the same concentration.
Changes in hydrogen ion concentration have no effect upon the read-
ing between pi 1 5.50 and 8.50 (determinations in steps of 0.2 pH units)
nor does the presence of 1 per cent HC1 alter the reading. In a like
manner oxygenation or reduction has no effect. This might be antici-
pated from the transmission characteristics of the green filter used.
The plasma introduced with the whole blood in preparing the dilute
solutions of hemoglobin constitutes a small but variable source of error.
This error appears to be partly the result of a slight degree of hemolysis
which takes place in the whole blood upon standing, though it is kept at
about 2° C. until used. The error varies between 1 and 4 per cent, and
as it is determinable, a correction can be applied. This factor, due to
pigments in the plasma, ma}- be determined by suspension of whole
380
12 15
18
21 24 27
30
Fir.. 4. Shift with time of the transmission of dilute hemoglobin solutions
(dilution 1 to 2,000). Ordinates represent readings of milliammeter ; abscissae,
time elapsed since the preparation of the hemoglobin solutions.
blood in isotonic salt solution in the same dilution as that used to pre-
pare the hemoglobin solution. The cells are then removed by centri-
fuging. The supernatant fluid is read by the pyrometer and compared
with the value for distilled water.
Another important factor influencing the determination in dilute
solutions of hemoglobin is the time elapsed since the preparation of the
solution. Readings were taken on samples of the same solution at a
ics of time intervals after its preparation. In Fig. 4 the results of
two such experiments are plotted. Between the time of preparation of
tin- solution and 4 to 8 hours later the transmission increases, which
causes an apparent decrease in the hemoglobin concentration; while be-
twi-en 10 nnd 48 hours the reading remains constant. The variation in
the final equilibrium attained in the case of these two different samples
HEMOGLOBIN CONCENTRATION IN DILUTE SOLUTIONS 523
of blood represents a difference of 1 per cent in hemoglobin concentra-
tion. This change in the readings with time may represent the trans-
formation of at least a part of the oxyhemoglobin into some other more
stable form. This shift with time makes it necessary that all readings
be taken only after the attainment of the final constant transmission
value. It has therefore been customary to make readings between 18
and 24 hours after the preparation of the solutions.
To determine the accuracy of the method a number of calibration
curves were made for different blood samples. A series of solutions
were prepared by dilution of the original 1 : 2,000 solution. These solu-
tions were read by the pyrometer. A calibration curve was then con-
structed from these values. Care was taken to control the factors of
salt content, plasma error, the time factor and errors of the optical sys-
4&0
460
440
420
400
380
360
0
£5
\
X^
1
\
\*\
\,
\
V
\
\-
\
\%
B>
\:\
•
\
\
\
V
10 20 30 40 50 60 70 80 90 1C
FIG. 5. Two typical calibration curves. Ordinates represent readings of mil-
liammeter; abscissae, per cent concentration of hemoglobin, 100 per cent being
equivalent to a 1 to 2,000 dilution of whole blood (ox).
tem. Two typical calibration curves are illustrated in Fig. 5. Other
hemoglobin solutions prepared by appropriate dilution of the original
1 : 2,000 solution were read in the optical pyrometer and their concentra-
tion determined by reference to the calibration curve for the same blood
sample. A number of such determinations have amply demonstrated
that the maximum error is 1 per cent. (Table I.)
524 A. K. PAR I 'ART. W. R. AMBERSOX, AND D. R. STEWART
TABLE I
Data on the Accuracy of the Method
I* -nt Xo.
Percentage of
Hemoglobin by
Dilution
Percentage of
Hemoglobin liy
Pyrometer
Percentage Error
1
2.5
2.0
-0.5
2
5.0
4.6
-O.I
3
10.0
10.8
+0.8
4
20.0
19.0
-1.0
5
25.0
25.0
0
6
37.5
38.0
+0.5
7
40.0
40.9
+0.9
8
50.0
51.0
+ 1.0
9
75.0
74.3
-0.7
10
80.0
80.5
+0.5
SUMMARY
The optical pyrometer has heen used to determine hemoglobin con-
centration in dilute solutions with a maximum error of 1 per cent. Ac-
curate determinations can he made with ease and rapidity. The out-
standing features of the apparatus are simplicity of construction and a
high degree of sensitivity.
Acknowledgment: \Ye wish to thank Dr. D. T ,. Drahkin for the
suggestion which led to the investigation of the shift with time in dilute
hemoglobin solutions.
BIBLIOGRAPHY
AMBERSON, W. R., 1922. Jour. Gen. Physio!., 4: 517.
FOWWEATHER, F. S.. ](>2<>. ttiochcw. Jour.. 20: 9.3.
PARPART, A. K., 1931. /</,•/. Hull.. 61: 500.
SCHWENTKER, F. F., 1929. Jour. Lab. and din. Mcd., 15: 247.
VAN SLVKE, D. D., AND A. KILLER, 1928. Jour. Biol. Clictn.. 78: 807.
INDEX
A CTIVATION, total, as related to
cleavage in artificially activated
Urechis eggs, 45.
ADOLPH, EDWARD F. Body size as a
factor in the metamorphosis of tad-
poles, 376.
— , - — . The size of the body and
the size of the environment in the
growth of tadpoles, 350.
Adrenaline, action in elasmobranch
fishes, 93.
ALEXANDER, GORDON. The significance
of hydrogen ion concentration in the
biology of Euglena gracilis Klebs,
165.
AMBERSON, W. R. See Parpart, Amber-
son and Stewart, 518.
Amphiuma tridactyla, oxygen and car-
bon dioxide transport by blood, 211.
Arbacia, surface tension of eggs, 273.
Asterias eggs, oxygen consumption rate
before and after fertilization, 468.
gAUMBERGER, J. P. and L. Mi-
CHAELIS. The blood pigments of
Urechis caupo, 417.
Blood, of Amphiuma, oxygen and carbon
dioxide transport, 211.
, of marine fishes, respiratory func-
tion, 427.
, of sea lion, respiratory function,
422.
— , respiratory function, in Urechis
caupo, 185.
Blood flukes, eggs of, effect of environ-
mental factors on development and
hatching, 120.
Blood pigments, of Urechis caupo, 417.
BLUM, H. F. and G. C. McBRiDE. Stud-
ies of photodynamic action, III, 316.
BURKENROAD, M. D. A new pentamer-
ous hydromedusa from the Tortugas,
115.
QARBON dioxide transport by blood
of Amphiuma, 211.
Carassius auratus, melanophores in ex-
perimental wounds, 73.
CAROTHERS, E. ELEANOR. The matura-
tion divisions and segregation of
heteromorphic homologous chromo-
somes in Acrididae (Orthoptera),
324.
Cellulose, digestion of, by termites, 85.
Cercaria parvicaudata, n. sp., 254.
Cercaria sensifera, n. sp., 259.
Chromosomes of domestic turkey, 157.
Citellus tridecemlineatus pallidus, Allen,
laboratory reproduction studies, 101.
Cleavage, as related to total activation
in artificially cultivated eggs of
Urechis, 45.
COE, WESLEY R. Spermatogenesis in
California oyster, 309.
Cytoplasmic contraction in Difflugia,
223.
DEVELOPMENTAL axis, determina-
tion of, in Fucus eggs, 294.
DICKMAN, ALBERT. Studies on the in-
testinal flora of termites with refer-
ence to their ability to digest cellu-
lose, 85.
Difflugia, movement and response, 223.
Digestion, of cellulose, by termites, and
their intestinal flora, 85.
Dilution of sea water, effect on activity
and longevity of marine cercariae,
242.
Diploiclism, genetic evidence for, of bi-
parental males in Habrobracon, 139.
— , in male parts in gynandromorphs
of Habrobracon, 478.
Drosophila, mutation rate, as affected
by continuous and interrupted irra-
diation, 133.
"PLASMOBRANCH fishes, innervation
of stomach and rectum and action
of adrenaline, 93.
Environment, effect on development and
hatching of blood fluke eggs, 120.
— , size of, as affecting body size of
tadpoles, 350.
Euglena gracilis Klebs, effect of hydro-
gen ion concentration on growth,
387.
— , significance of hydrogen
ion concentration in biology of, 165.
525
526
INDEX
pAU.KNMK. G. H. Notes on the
hanism and on intestinal
ir.it ion in ( "haetopterus variope-
us, 47J.
Feeding mechanism, Chaetopterus vario-
jicdatus, 472.
I I MRKiN, M. See Redfield and Florkin,
185.
FLORKIN, M., and A. C. REDFIELD. On
the respiratory function of the blood
of the sea lion, 422.
Fucus, eggs, influence in determination
of developmental axis, 294.
("""OLDFISH, occurrence of melano-
phores in experimental wounds, 73.
Ground squirrel, laboratory reproduction
studies, 101.
Growth, of Euglena gracilis Klebs, effect
of hydrogen ion concentration, 387.
— , of tadpoles, as affected by size of
environment, 350.
< .ynandromorph, of Habrobracon from
post-reduced binucleate egg, 481.
— , of Habrobracon, diploid male
parts, 478.
UABROBRACON, diploid male parts
in gynandromorphs of, 478.
— , genetic evidence for diploidism in
liiparental males, 139.
— , gynandromorph from post-reduced
binucleate egg, 481.
HALL, F. G. The respiration of puffer
fish, 457.
HALL, VICTOR E. The muscular activity
and oxygen consumption of Urechis
caupo, 400.
HARVEY, E. XKWTON. See Taylor and
Harvey, 280.
— , - — . The tension at the sur-
face of marine eggs, espc< iallv those
of the sea urchin, Arbacia, 273.
Hatching, of blood fluke eggs, effect of
environmental factors, 120.
Hemoglobin concentration, determina-
tion of, in dilute solutions, 5 IS.
Hemolysis, photodynamic, and by non-
irradiated cosine, difference in mech-
ani-m, .^ 1 <>.
II imi concentration, effect on
growth of Euglena gracilis Kleb-,
387.
— , significance in biology of
Li ili- Klebs, 165.
Ilydroniedu^a, pentamerous, from Tor-
tugas, 115.
TNNERVATION of stomach and rec-
tum in elasmobranch fishes, 93.
Intestinal flora of termites, and digestion
of cellulose, 85.
Irradiation, continuous and interrupted,
effects on mutation rate of Droso-
phila, 133.
JACOBS, M. H. See Williams and
J Jacobs, 485.
JAHN, THEO. L. Studies on the physi-
ology of the euglenoid flagellates,
III, 387.
JOHNSON, GEORGE E. and NELSON J.
\YADE. Laboratory reproduction
studies on the ground squirrel, Citel-
lus tridecemlineatus pallidus, Allen,
101.
T IGHT responses of parasitic water
mites, influence of host, 497.
Longevity, certain marine cercariae, as
affected by dilution of sea water,
242.
Luxz, BRENTON R. The innervation of
the stomach and rectum and the
action of adrenaline in elasmobranch
fishes, 93.
V/f ARINE Biological Laboratory, thir-
ty-third report, 1.
MAM, S. ( ). Movement and response in
Dilllugia with special reference to
the nature of cytoplasmic contrac-
tion, 223.
Maturation divisions in Acrididae, 324.
MiT.kim . G. C. See Blum and Mc-
Ihide, 316.
Melanophores, in experimental wounds
of goldfish, T.v
Metamorphosis of tadpoles, body size as
factor in, 376.
MKIIAELIS, L. See Baumberger and
Midlaeli-, 417.
Mitogenetic radiation, theory of, 280.
Muscular activity and oxygen consump-
tion of Urechis caupo, 400.
Mutation, rate in Drosophila, effects of
continuous and interrupted irradia-
tion, l.v>.
QNOKATO, A. R. and II. \V. STUN-
KAKU. Tlic effect of certain environ-
mental factors on the development
and hatching of the eggs of blood
flukes, 120.
INDEX
527
Osmotic hemolysis, question as to
whether it is an all-or-none phe-
nomenon, 500.
Ostrea lurida, spermatogenesis in, 309.
Oxygen consumption and muscular ac-
tivity, Urechis caupo, 400.
— , rate of, by Asterias eggs be-
fore and after fertilization, 468.
Oxygen transport, by blood of Amphi-
uma, 211.
Oyster, California, spermatogenesis, 309.
pARPART, A. K. Is osmotic hemoly-
sis an all-or-none phenomenon? 500.
— , - — , W. R. AMBERSON and D. R.
STEWART. The determination of
hemoglobin concentration in dilute
solutions, 518.
PATTERSON, J. T. Continuous versus
interrupted irradiation and the rate
of mutation in Drosophila, 133.
Photodynamic action, difference in mech-
anism between photodynamic hemo-
lysis and hemolysis by irradiated
eosine, 316.
Puffer fish, respiration of, 457.
IJ ECTUM, innervation of, elasmo-
branch fishes, 93.
REDFIELD, A. C. See Redfield and Flor-
kin, 422.
— , - — . The respira-
tory function of the blood of Urechis
caupo, 185.
Reproduction studies, in the laboratory,
on the ground squirrel, 101.
Respiration, blood pigments of Urechis
caupo, 417.
— , function of blood of marine fishes,
427.
— , function of blood of sea lion, 422.
— , function of Urechis caupo, 185.
— , intestinal, in Chaetopterus vario-
pedatus, 472.
— , muscular activity and oxygen con-
sumption of Urechis caupo, 400.
— , oxygen and carbon dioxide trans-
port by blood of Amphiuma tridac-
tyla, 211.
— , oxygen consumption rate of As-
terias eggs before and after fertili-
zation, 468.
Response, in Difflugia, 223.
ROOT, R. W. The respiratory function
of the blood of marine fishes, 427.
gCOTT, W. J. Oxygen and carbon di-
oxide transport by the blood of the
urodele, Amphiuma tridactyla, 211.
Sea lion, respiratory function of blood,
422.
Sea water, dilution of, effect on activity
and longevity of marine cercariae,
242.
Segregation of hetcromorphic homologous
chromosomes in Acrididae, 324.
Size, of body and of environment in
growth of tadpoles, 350.
Size, of body in metamorphosis of tad-
poles, 376.
SMITH, GEORGE MILTON. The occur-
rence of melanophores in certain ex-
perimental wounds of the goldfish
(Carassius auratus), 73.
Sodium chloride, physiological differences
between different preparations, 485.
Spermatogenesis, of California oyster,
309.
STANCATI, M. F. See Whiting and Stan-
cati, 478.
STEWART, D. R. See Parpart, Amberson
and Stewart, 518.
Stomach, innervation of, elasmobranch
fishes, 93.
STUNKARD, H. W. See Onorato and
Stunkard, 120.
— , - — . The effect of dilution of sea
water on the activity and longevity
of certain marine cercariae, 242.
Surface tension marine eggs, especially
Arbacia, 273.
'"PADPOLES, body size and environ-
ment size, 350.
— , body size in metamorphosis of, 376.
TANG, PEI-SUNG. The rate of oxygen
consumption of Asterias eggs before
and after fertilization, 468.
TAYLOR, G. WELLFORD and E. NEWTON
HARVEY. The theory of mitogenetic
radiation, 280.
Termites, their intestinal flora with ref-
erence to digestion of cellulose, 85.
Thirty-third report of the Marine Bio-
logical Laboratory, 1.
TORVIK, M. M. Genetic evidence for
diploidism of biparental males in
Habrobracon, 139.
Turkey, domestic, chromosomes of, 157.
TYLER, ALBERT. The relation between
cleavage and total activation in arti-
528
IXDEX
finally cultivated eggs of Urechis,
45.
TTKl.i HIS, relation between cleavage
ami total activation in artificially
cultivated eggs, 45.
his caupo, blood pigments of, 417.
— , muscular activity and oxygen
consumption, 400.
-, respiratory function of blood,
185.
AA/ADE, NELSON J. See Johnson and
Wade, 101.
Water mites, parasitic, influence of host
on light responses, 497.
WELSH, JOHN H. Specific influence of
the host on the light responses of
parasitic water mites, 497.
WERNER, ORILLA STOTLER. The chro-
mosomes of the domestic turkey,
157.
WHITAKER, D. M. Some observations
on the eggs of Fucus and upon their
mutual influence in the determina-
tion of the developmental axis, 294.
WHITING, P. W. Diploid male parts in
gynandromorphs of Habrobracon,
478.
WHITING, P. W. and M. F. STANCATI.
A gynandromorph of Habrobracon
from a post-reduced binucleate egg,
481.
WILLIAMS, MARY MORRISON and M. H.
JACOBS. On certain physiological
differences between different prepa-
rations of so-called "chemically
pure" sodium chloride, 485.
Volume LXI
Number 1
THE
BIOLOGICAL BULLETIN
PUBLISHED BY
THE MARINE BIOLOGICAL LABORATO
Editorial Board
GARY N. CALKINS, Columbia University
E. G. CONKLIN, Princeton University
E. N. HARVEY, Princeton University
SELIG HECHT, Columbia University
M. H. JACOBS, University of Pennsylvania
H. S. JENNINGS, Johns Hopkins University
E. E. JUST, Howard University
FRANK R. LILLIE, University of Chicago
CARL R. MOORE, University of Chicago
GEORGE T. MOORE, Missouri Botanical Garden
T. H. MORGAN, California Institute of Technology
G. H. PARKER, Harvard University
W. M. WHEELER, Harvard University
EDMUND B. WILSON, Columbia University
ALFRED C. REDFIELD, Harvard University
Managing Editor
AUGUST, 1931
Printed and Issued by
LANCASTER PRESS, Inc.
PRINCE 8C LEMON STS.
LANCASTER, PA.
THE BIOLOGICAL BUI.I.KTLX is issued six times a year. Single
numbers, $1.75. Subscription per volume (3 numbers), $4 50.
Subscriptions and other matter should be addressed to the
Biological Bulletin, Prince and Lemon Streets, Lancaster, Pa.
Agent for Great Britain: \Yheldon & Wesley, Limited, 2, 3 and
4 Arthur Street, New Oxford Street, London, W.C. 2.
Communications relative to manuscripts should be sent to
the Managing Editor, Marine Biological Laboratory, Woods
Hole, Mass., between May 1 and November 1 and to the
Zoological Laboratory, Harvard University, Cambridge, Mass.,
during the remainder of the year.
Entered October in, 1902, at Lancaster, Pa., as second-class matter under
Act of Congress of July 16, 1894.
CONTENTS
Page
THIRTY-THIRD REPORT OF THE MARINE BIOLOGICAL LABO-
RATORY l
TYLER, ALBERT
The Relation between Cleavage and Total Activation in
Artificially Activated Eggs of Urechis 45
SMITH, GEORGE MILTON
The Occurrence of Melanophores in certain Experimental
Wounds of the Goldfish (Carassius auratus) 73
DICKMAN, ALBERT
Studies on the Intestinal Flora of Termites with reference
to their Ability to Digest Cellulose 85
LUTZ, BRENTON R.
The Innervation of the Stomach and Rectum and the Action
of Adrenaline in Elasmobranch Fishes 93
JOHNSON, GEORGE E., AND NELSON J. WADE
Laboratory Reproduction Studies on the Ground Squirrel,
Citellus tridecemlineatus pallidus, Allen 101
BURKENROAD, M.D.
A New Pentamerous Hydromedusa from the Tortugas 115
ONORATO, A. R., AND H. W. STUNKARD
The Effect of certain Environmental Factors on the Develop-
ment and Hatching of the Eggs of Blood Flukes 120
Volume LXI Number 2
THE
BIOLOGICAL BULLETIN
PUBLISHED BY
THE MARINE BIOLOGICAL LABORATORY
Editorial Board
GARY N. CALKINS, Columbia University
E. G. CONKLIN, Princeton University FRANK R. LlLLIE, University of Chicago
E. N. HARVEY, Princeton University CARL R. MOORE, University of Chicago
SELIG HECHT, Columbia University GEORGE T. MOORE, Missouri Botanical Garden
LEIGH HOADLEY, Harvard University T. H. MORGAN, California Institute of Technology
M. H. JACOBS, University of Pennsylvania G. H. PARKER, Harvard University
H. S. JENNINGS, Johns Hopkins University W. M. WHEELER, Harvard University
E. E. JUST, Howard University EDMUND B. WILSON, Columbia University
ALFRED C. REDFLELD, Harvard University
Managing Editor
OCTOBER, 1931
Printed and Issued by
LANCASTER PRESS, Inc.
PRINCE & LEMON STS.
LANCASTER, PA.
THE BIOLOGICAL BULLETIN is issued six times a year. Single
numbers, $1.75. Subscription per volume (3 numbers), $4.50.
Subscriptions and other matter should be addressed to the
Biological Bulletin, Prince and Lemon Streets, Lancaster, Pa.
Agent for Great Britain: Wheldon & Wesley, Limited, 2, 3 and
4 Arthur Street, New Oxford Street, London, W.C. 2.
Communications relative to manuscripts should be sent to
the Managing Editor, Marine Biological Laboratory, Woods
Hole, Mass., between May 1 and November 1 and to the
Zoological Laboratory, Harvard University, Cambridge, Mass.,
during the remainder of the year.
Entered October 10, 1902, at Lancaster, Pa., as second-class matter under
Act of Congress of July 16, 1894.
CONTENTS
Page
PATTERSON, J. T.
Continuous versus Interrupted Irradiation and the Rate of
Mutation in Drosophila 133
TORVIK, M. M.
Genetic Evidence for Diploidism of Biparental Males in
Habrobracon 139
WERNER, ORILLA STOTLER
The Chromosomes of the Domestic Turkey 157
ALEXANDER, GORDON
The Significance of Hydrogen Ion Concentration in the
Biology of Euglena gracilis Klebs 165
REDFIELD, A. C., AND M. FLORKIN
The Respiratory Function of the Blood of Urechis caupo .... 185
SCOTT, W. J.
Oxygen and Carbon Dioxide Transport by the Blood of the
Urodele, Amphiuma tridactyla 211
MAST, S. O.
Movement and Response in Difflugia with special reference
to the Nature of Cytoplasmic Contraction 223
STUNKARD, H. W.
The Effect of Dilution of Sea Water on the Activity and
Longevity of Certain Marine Cercariae .... 242
Volume LXI Number 3
THE
BIOLOGICAL BULLETIN
PUBLISHED BY
THE MARINE BIOLOGICAL LABORATORY
Editorial Board
GARY N. CALKINS, Columbia University
E. G. CONKLIN, Princeton University FRANK R. LlLLIE, University of Chicago
E. N. HARVEY, Princeton University CARL R. MOORE, University of Chicago
SELIG HECHT, Columbia University GEORGE T. MOORE, Missouri Botanical Garden
LEIGH HOADLEY, Harvard University T. H. MORGAN, California Institute of Technology
M. H. JACOBS, University of Pennsylvania G. H. PARKER, Harvard University
H. S. JENNINGS, Johns Hopkins University W. M. WHEELER, Harvard University
E. E. JUST, Howard University EDMUND B. WILSON, Columbia University
ALFRED C. REDFIELD, Harvard University
Managing Editor
DECEMBER, 1931
Printed and Issued by
LANCASTER PRESS, Inc.
PRINCE & LEMON STS.
LANCASTER, PA.
THE BIOLOGICAL BULLKTIX is issued six times a year. Single
numbers, $1.75. Subscription per volume (3 numbers), $4.50.
Subscriptions and other matter should be addressed to the
Biological Bulletin, Prince and Lemon Streets, Lancaster, Pa.
Agent for Great Britain: \Yheldon & Wesley, Limited, 2, 3 and
4 Arthur Street, New Oxford Street, London, \Y.C. 2.
Communications relative to manuscripts should be sent to
the Managing Editor, Marine Biological Laboratory, Woods
Hole, Mass., between May 1 and November 1 and to the
Institute of Biology, Divinity Avenue, Cambridge, Mass., during
the remainder of the year.
Kntcrr.l i )( -inher 10, 1902, at Lancaster, Pa., as second-class matter under
Act of Congress of July 16, 1894.
CONTENTS
HARVEY, E. NEWTON Page
The Tension at the Surface of Marine Eggs, especially those of the
Sea Urchin, Arbacia 273
TAYLOR, G. WELLFORD, AND E. NEWTON HARVEY
The Theory of Mitogenetic Radiation 280
WHITAKER, D. M.
Some Observations on the Eggs of Fucus and upon their Mutual
Influence in the Determination of the Developmental Axis 294
COE, WESLEY R.
Spermatogenesis in the California Oyster (Ostrea lurida) 309
BLUM, H. F., AND G. C. MCBRIDE
Studies of Photodynamic Action, III. The difference in mechanism
between photodynamic hemolysis and hemolysis by non-irradiated
eosine 316
CAROTHERS, E. ELEANOR
The Maturation Divisions and Segregation of Heteromorphic Homol-
ogous Chromosomes in Acrididae (Orthoptera) 324
ADOLPH, EDWARD F.
The Size of the Body and the Size cf the Environment in the Growth
of Tadpoles 350
ADOLPH, EDWARD F.
Body Size as a Factor in the Metamorphosis of Tadpoles 376
JAHN, THEO. L.
Studies on the Physiology of the Euglenoid Flagellates, III. The
effect of hydrogen ion concentration on the growth of Euglena gracilis
Klebs 387
HALL, VICTOR E.
The Muscular Activity and Oxygen Consumption of Urechis caupo . . 400
BAUMBERGER, J. P., AND L. MICHAELIS
The Blood Pigments of Urechis caupo 417
FLOFKIN, MARCEL, AND ALFRED C. REDFIELD
On the Respiratory Function of the Blood of the Sea Lion 422
ROOT, R. W.
The Respiratory Function of the Blood of Marine Fishes 427
HALL, F. G.
The Respiration of Puffer Fish. ... 457
TANG, PEI-SUNG
The Rate of Oxygen Consumption of Asterias Eggs Before and After
Fertilization 468
FAULKNER, G. H.
Notes on the Feeding Mechanism and on Intestinal Respiration in
Chaetopterus variopedatus 472
WHITING, P. W.
Diploid Male Parts in Gynandromcrphs of Habrobracon 478
WHITING, P. W., AND M. F. STANCATI
A Gynandromorph of Habrobracon from a Post-Reduced Binucleate
Egg. ... 481
WILLIAMS, MARY MORRISON, AND M. H. JACOBS
On Certain Physiological Differences between Different Preparations
of So-Called " Chemically Pure " Sodium Chloride 485
WELSH, JOHN H.
Specific Influence of the Host on the Light Responses of Parasitic
Water Mites 497
PARPART, ARTHUR K.
Is Osmotic Hemolysis an All-or-none Phenomenon? 500
PARPART, A. K., W. R. AMBERSON AND D. R. STEWART
The Determination of Hemoglobin Concentration in Dilute Solutions 518
* HI. WII01 LIBRARY
WH 17 ID
Live Music
Archive
Librivox
Free Audio
Metropolitan Museum
Cleveland
Museum of Art
Internet
Arcade
Console Living Room
Open Library
American
Libraries
TV News
Understanding
9/11